Dataset Viewer
Auto-converted to Parquet Duplicate
instruction
stringlengths
314
345
response
stringlengths
7.2k
8.46k
query
stringlengths
34
65
num_papers
int64
10
10
๋‹ค์Œ ์งˆ๋ฌธ์— ๋Œ€ํ•ด ๋…ผ๋ฌธ์„ ๊ฒ€์ƒ‰ํ•˜๊ณ  ๋žญํ‚นํ•˜์—ฌ ๊ฒฐ๊ณผ๋ฅผ ์ œ์‹œํ•˜์„ธ์š”. ์งˆ๋ฌธ: What are the advantages of RNN over traditional approaches? ์ถœ๋ ฅ ํ˜•์‹: 1. Evidence Table (Top-K): ๊ฒ€์ƒ‰๋œ ๋…ผ๋ฌธ์˜ ์ƒ์œ„ K๊ฐœ๋ฅผ ํ…Œ์ด๋ธ” ํ˜•์‹์œผ๋กœ ์ œ์‹œ 2. Ranking Breakdown: ๊ฐ ๋…ผ๋ฌธ์˜ ๋žญํ‚น ์ ์ˆ˜ ๊ตฌ์„ฑ ์š”์†Œ๋ฅผ ๊ณต๊ฐœ 3. Answer Summary: ๊ฒ€์ƒ‰ ๊ฒฐ๊ณผ๋ฅผ ์š”์•ฝํ•˜์—ฌ ์ œ์‹œ 4. Charts: ์—ฐ๋„๋ณ„ ํŠธ๋ Œ๋“œ, ์ €๋„ ๋ถ„ํฌ, ์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ ์ฐจํŠธ์— ๋Œ€ํ•œ ์„ค๋ช… ๋ชจ๋“  ์ฃผ์žฅ์€ Evidence Table์˜ ๋…ผ๋ฌธ์— ๊ทผ๊ฑฐํ•ด์•ผ ํ•˜๋ฉฐ, ๋žญํ‚น ๊ทผ๊ฑฐ๋ฅผ ๋ช…์‹œํ•ด์•ผ ํ•ฉ๋‹ˆ๋‹ค.
{ "evidence_table": [ { "rank": 1, "title": "A Deep Investigation of RNN and Self-attention for the Cyrillic-Traditional Mongolian Bidirectional Conversion", "authors": "Muhan Na, Rui Liu, Feilong", "year": 2022, "venue": "arXiv", "study_type": "experimental", "score": 0.5780555555555555, "url": "https://arxiv.org/abs/2209.11963v1" }, { "rank": 2, "title": "What is the best RNN-cell structure to forecast each time series behavior?", "authors": "Rohaifa Khaldi, Abdellatif El Afia, Raddouane Chiheb", "year": 2022, "venue": "arXiv", "study_type": "experimental", "score": 0.5558333333333333, "url": "https://arxiv.org/abs/2203.07844v2" }, { "rank": 3, "title": "Attention as an RNN", "authors": "Leo Feng, Frederick Tung, Hossein Hajimirsadeghi", "year": 2024, "venue": "arXiv", "study_type": "experimental", "score": 0.5513888888888888, "url": "https://arxiv.org/abs/2405.13956v2" }, { "rank": 4, "title": "Proceedings to the 27th Workshop \"What Comes Beyond the Standard Models\" Bled, July 8-17, 2024", "authors": "R. Bernabei, P. Belli, A. Bussolotti", "year": 2025, "venue": "arXiv", "study_type": "theoretical", "score": 0.5402777777777777, "url": "https://arxiv.org/abs/2504.17803v1" }, { "rank": 5, "title": "A One-Dimensional Energy Balance Model Parameterization for the Formation of CO2 Ice on the Surfaces of Eccentric Extrasolar Planets", "authors": "Vidya Venkatesan, Aomawa L. Shields, Russell Deitrick", "year": 2025, "venue": "arXiv", "study_type": "theoretical", "score": 0.5402777777777777, "url": "https://arxiv.org/abs/2501.11667v1" }, { "rank": 6, "title": "FlashRNN: I/O-Aware Optimization of Traditional RNNs on modern hardware", "authors": "Korbinian Pรถppel, Maximilian Beck, Sepp Hochreiter", "year": 2024, "venue": "arXiv", "study_type": "theoretical", "score": 0.5336111111111111, "url": "https://arxiv.org/abs/2412.07752v3" }, { "rank": 7, "title": "Gates Are Not What You Need in RNNs", "authors": "Ronalds Zakovskis, Andis Draguns, Eliza Gaile", "year": 2021, "venue": "arXiv", "study_type": "experimental", "score": 0.5269444444444444, "url": "https://arxiv.org/abs/2108.00527v3" }, { "rank": 8, "title": "What can a cook in Italy teach a mechanic in India? Action Recognition Generalisation Over Scenarios and Locations", "authors": "Chiara Plizzari, Toby Perrett, Barbara Caputo", "year": 2023, "venue": "arXiv", "study_type": "theoretical", "score": 0.5136111111111111, "url": "https://arxiv.org/abs/2306.08713v2" }, { "rank": 9, "title": "The Galactic Chemical Evolution of phosphorus observed with IGRINS", "authors": "G. Nandakumar, N. Ryde, M. Montelius", "year": 2022, "venue": "arXiv", "study_type": "experimental", "score": 0.5069444444444444, "url": "https://arxiv.org/abs/2210.04940v1" }, { "rank": 10, "title": "PRAXA: A Framework for What-If Analysis", "authors": "Sneha Gathani, Kevin Li, Raghav Thind", "year": 2025, "venue": "arXiv", "study_type": "theoretical", "score": 0.5047222222222222, "url": "https://arxiv.org/abs/2510.09791v2" } ], "ranking_breakdown": [ { "paper_id": "2209.11963", "title": "A Deep Investigation of RNN and Self-attention for the Cyrillic-Traditional Mongolian Bidirectional ", "total_score": 0.5780555555555555, "query_match": 0.38888888888888884, "recency": 0.7, "study_type_priority": 0.7, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2203.07844", "title": "What is the best RNN-cell structure to forecast each time series behavior?", "total_score": 0.5558333333333333, "query_match": 0.3333333333333333, "recency": 0.7, "study_type_priority": 0.7, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2405.13956", "title": "Attention as an RNN", "total_score": 0.5513888888888888, "query_match": 0.2222222222222222, "recency": 0.9, "study_type_priority": 0.7, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2504.17803", "title": "Proceedings to the 27th Workshop \"What Comes Beyond the Standard Models\" Bled, July 8-17, 2024", "total_score": 0.5402777777777777, "query_match": 0.24444444444444444, "recency": 1.0, "study_type_priority": 0.5, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2501.11667", "title": "A One-Dimensional Energy Balance Model Parameterization for the Formation of CO2 Ice on the Surfaces", "total_score": 0.5402777777777777, "query_match": 0.24444444444444444, "recency": 1.0, "study_type_priority": 0.5, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2412.07752", "title": "FlashRNN: I/O-Aware Optimization of Traditional RNNs on modern hardware", "total_score": 0.5336111111111111, "query_match": 0.2777777777777778, "recency": 0.9, "study_type_priority": 0.5, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2108.00527", "title": "Gates Are Not What You Need in RNNs", "total_score": 0.5269444444444444, "query_match": 0.3111111111111111, "recency": 0.6, "study_type_priority": 0.7, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2306.08713", "title": "What can a cook in Italy teach a mechanic in India? Action Recognition Generalisation Over Scenarios", "total_score": 0.5136111111111111, "query_match": 0.2777777777777778, "recency": 0.8, "study_type_priority": 0.5, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2210.04940", "title": "The Galactic Chemical Evolution of phosphorus observed with IGRINS", "total_score": 0.5069444444444444, "query_match": 0.21111111111111108, "recency": 0.7, "study_type_priority": 0.7, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2510.09791", "title": "PRAXA: A Framework for What-If Analysis", "total_score": 0.5047222222222222, "query_match": 0.15555555555555556, "recency": 1.0, "study_type_priority": 0.5, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 } ], "answer_summary": "๊ฒ€์ƒ‰ ์งˆ๋ฌธ: What are the advantages of RNN over traditional approaches?\n\n๊ฒ€์ƒ‰ ๊ฒฐ๊ณผ ์š”์•ฝ:\n- ์ด 10๊ฐœ์˜ ๋…ผ๋ฌธ์ด ๊ฒ€์ƒ‰๋˜์—ˆ์Šต๋‹ˆ๋‹ค.\n- ์†Œ์Šค๋ณ„ ๋ถ„ํฌ: arxiv(10)\n- ์—ฐ๋„ ๋ฒ”์œ„: 2021-2025\n- ํ‰๊ท  ๋žญํ‚น ์ ์ˆ˜: 0.535\n\n์ฃผ์š” ๋ฐœ๊ฒฌ:\n- ์ƒ์œ„ ๋…ผ๋ฌธ๋“ค์€ ์ฃผ๋กœ arXiv, arXiv, arXiv์—์„œ ๋ฐœํ‘œ๋˜์—ˆ์Šต๋‹ˆ๋‹ค.\n- ์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ: experimental(5), theoretical(5)", "charts": { "year_trend": "์—ฐ๋„๋ณ„ ํŠธ๋ Œ๋“œ ์ฐจํŠธ", "venue_distribution": "์ €๋„ ๋ถ„ํฌ ์ฐจํŠธ", "study_type_distribution": "์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ ์ฐจํŠธ" } }
What are the advantages of RNN over traditional approaches?
10
๋‹ค์Œ ์งˆ๋ฌธ์— ๋Œ€ํ•ด ๋…ผ๋ฌธ์„ ๊ฒ€์ƒ‰ํ•˜๊ณ  ๋žญํ‚นํ•˜์—ฌ ๊ฒฐ๊ณผ๋ฅผ ์ œ์‹œํ•˜์„ธ์š”. ์งˆ๋ฌธ: What are the technical details of RNN? ์ถœ๋ ฅ ํ˜•์‹: 1. Evidence Table (Top-K): ๊ฒ€์ƒ‰๋œ ๋…ผ๋ฌธ์˜ ์ƒ์œ„ K๊ฐœ๋ฅผ ํ…Œ์ด๋ธ” ํ˜•์‹์œผ๋กœ ์ œ์‹œ 2. Ranking Breakdown: ๊ฐ ๋…ผ๋ฌธ์˜ ๋žญํ‚น ์ ์ˆ˜ ๊ตฌ์„ฑ ์š”์†Œ๋ฅผ ๊ณต๊ฐœ 3. Answer Summary: ๊ฒ€์ƒ‰ ๊ฒฐ๊ณผ๋ฅผ ์š”์•ฝํ•˜์—ฌ ์ œ์‹œ 4. Charts: ์—ฐ๋„๋ณ„ ํŠธ๋ Œ๋“œ, ์ €๋„ ๋ถ„ํฌ, ์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ ์ฐจํŠธ์— ๋Œ€ํ•œ ์„ค๋ช… ๋ชจ๋“  ์ฃผ์žฅ์€ Evidence Table์˜ ๋…ผ๋ฌธ์— ๊ทผ๊ฑฐํ•ด์•ผ ํ•˜๋ฉฐ, ๋žญํ‚น ๊ทผ๊ฑฐ๋ฅผ ๋ช…์‹œํ•ด์•ผ ํ•ฉ๋‹ˆ๋‹ค.
{ "evidence_table": [ { "rank": 1, "title": "Deep learning in surgical process modeling: A systematic review of workflow recognition.", "authors": "Liu Zhenzhong, Chen Kelong, Wang Shuai", "year": 2025, "venue": "J Biomed Inform", "study_type": "systematic_review", "score": 0.6271428571428572, "url": "https://pubmed.ncbi.nlm.nih.gov/39832608/" }, { "rank": 2, "title": "A hybrid deep learning framework for early detection of diabetic retinopathy using retinal fundus images.", "authors": "Sushith Mishmala, Sathiya A, Kalaipoonguzhali V", "year": 2025, "venue": "Sci Rep", "study_type": "experimental", "score": 0.6042857142857143, "url": "https://pubmed.ncbi.nlm.nih.gov/40307328/" }, { "rank": 3, "title": "Overcoming the Barrier of Incompleteness: A Hyperspectral Image Classification Full Model.", "authors": "Yang Jiaqi, Du Bo, Zhang Liangpei", "year": 2024, "venue": "IEEE Trans Neural Netw Learn Syst", "study_type": "experimental", "score": 0.5978571428571429, "url": "https://pubmed.ncbi.nlm.nih.gov/37279129/" }, { "rank": 4, "title": "ElectroPhysiomeGAN: Generation of Biophysical Neuron Model Parameters from Recorded Electrophysiological Responses.", "authors": "Kim Jimin, Peng Minxian, Chen Shuqi", "year": 2025, "venue": "bioRxiv", "study_type": "experimental", "score": 0.5892857142857143, "url": "https://pubmed.ncbi.nlm.nih.gov/40501917/" }, { "rank": 5, "title": "Precision diagnosis of burn injuries using imaging and predictive modeling for clinical applications.", "authors": "Rangaiah Pramod K B, Kumar B P Pradeep, Huss Fredrik", "year": 2025, "venue": "Sci Rep", "study_type": "experimental", "score": 0.5871428571428572, "url": "https://pubmed.ncbi.nlm.nih.gov/40038450/" }, { "rank": 6, "title": "Integration of RNN and CatBoost models in a tea-waste biochar filtration system for toxic organic pollutant removal efficiency prediction.", "authors": "Jha Stuti, Gaur Rama, Shahabuddin Syed", "year": 2025, "venue": "RSC Adv", "study_type": "theoretical", "score": 0.5607142857142857, "url": "https://pubmed.ncbi.nlm.nih.gov/40746784/" }, { "rank": 7, "title": "A recurrent neural network for soft sensor development using CHO stable pools in fed-batch process for SARS-CoV-2 spike protein production as a vaccine antigen.", "authors": "Reyes Sebastian-Juan, Voyer Robert, Durocher Yves", "year": 2025, "venue": "Biotechnol Prog", "study_type": "experimental", "score": 0.5585714285714285, "url": "https://pubmed.ncbi.nlm.nih.gov/40454837/" }, { "rank": 8, "title": "What's behind tight deadlines? Business causes of technical debt", "authors": "Rodrigo Rebouรงas de Almeida, Christoph Treude, Uirรก Kulesza", "year": 2021, "venue": "arXiv", "study_type": "survey", "score": 0.5539285714285714, "url": "https://arxiv.org/abs/2104.09330v5" }, { "rank": 9, "title": "Proceedings to the 27th Workshop \"What Comes Beyond the Standard Models\" Bled, July 8-17, 2024", "authors": "R. Bernabei, P. Belli, A. Bussolotti", "year": 2025, "venue": "arXiv", "study_type": "theoretical", "score": 0.5510714285714285, "url": "https://arxiv.org/abs/2504.17803v1" }, { "rank": 10, "title": "A One-Dimensional Energy Balance Model Parameterization for the Formation of CO2 Ice on the Surfaces of Eccentric Extrasolar Planets", "authors": "Vidya Venkatesan, Aomawa L. Shields, Russell Deitrick", "year": 2025, "venue": "arXiv", "study_type": "theoretical", "score": 0.5510714285714285, "url": "https://arxiv.org/abs/2501.11667v1" } ], "ranking_breakdown": [ { "paper_id": "39832608", "title": "Deep learning in surgical process modeling: A systematic review of workflow recognition.", "total_score": 0.6271428571428572, "query_match": 0.24285714285714283, "recency": 1.0, "study_type_priority": 0.9, "evidence_completeness": 0.9999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "40307328", "title": "A hybrid deep learning framework for early detection of diabetic retinopathy using retinal fundus im", "total_score": 0.6042857142857143, "query_match": 0.2857142857142857, "recency": 1.0, "study_type_priority": 0.7, "evidence_completeness": 0.9999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "37279129", "title": "Overcoming the Barrier of Incompleteness: A Hyperspectral Image Classification Full Model.", "total_score": 0.5978571428571429, "query_match": 0.3571428571428571, "recency": 0.9, "study_type_priority": 0.7, "evidence_completeness": 0.8999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "40501917", "title": "ElectroPhysiomeGAN: Generation of Biophysical Neuron Model Parameters from Recorded Electrophysiolog", "total_score": 0.5892857142857143, "query_match": 0.2857142857142857, "recency": 1.0, "study_type_priority": 0.7, "evidence_completeness": 0.8999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "40038450", "title": "Precision diagnosis of burn injuries using imaging and predictive modeling for clinical applications", "total_score": 0.5871428571428572, "query_match": 0.24285714285714283, "recency": 1.0, "study_type_priority": 0.7, "evidence_completeness": 0.9999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "40746784", "title": "Integration of RNN and CatBoost models in a tea-waste biochar filtration system for toxic organic po", "total_score": 0.5607142857142857, "query_match": 0.3142857142857143, "recency": 1.0, "study_type_priority": 0.5, "evidence_completeness": 0.8999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "40454837", "title": "A recurrent neural network for soft sensor development using CHO stable pools in fed-batch process f", "total_score": 0.5585714285714285, "query_match": 0.1714285714285714, "recency": 1.0, "study_type_priority": 0.7, "evidence_completeness": 0.9999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "2104.09330", "title": "What's behind tight deadlines? Business causes of technical debt", "total_score": 0.5539285714285714, "query_match": 0.42857142857142855, "recency": 0.6, "study_type_priority": 0.6, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2504.17803", "title": "Proceedings to the 27th Workshop \"What Comes Beyond the Standard Models\" Bled, July 8-17, 2024", "total_score": 0.5510714285714285, "query_match": 0.2714285714285714, "recency": 1.0, "study_type_priority": 0.5, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2501.11667", "title": "A One-Dimensional Energy Balance Model Parameterization for the Formation of CO2 Ice on the Surfaces", "total_score": 0.5510714285714285, "query_match": 0.2714285714285714, "recency": 1.0, "study_type_priority": 0.5, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 } ], "answer_summary": "๊ฒ€์ƒ‰ ์งˆ๋ฌธ: What are the technical details of RNN?\n\n๊ฒ€์ƒ‰ ๊ฒฐ๊ณผ ์š”์•ฝ:\n- ์ด 10๊ฐœ์˜ ๋…ผ๋ฌธ์ด ๊ฒ€์ƒ‰๋˜์—ˆ์Šต๋‹ˆ๋‹ค.\n- ์†Œ์Šค๋ณ„ ๋ถ„ํฌ: pubmed(7), arxiv(3)\n- ์—ฐ๋„ ๋ฒ”์œ„: 2021-2025\n- ํ‰๊ท  ๋žญํ‚น ์ ์ˆ˜: 0.578\n\n์ฃผ์š” ๋ฐœ๊ฒฌ:\n- ์ƒ์œ„ ๋…ผ๋ฌธ๋“ค์€ ์ฃผ๋กœ J Biomed Inform, Sci Rep, IEEE Trans Neural Netw Learn Syst์—์„œ ๋ฐœํ‘œ๋˜์—ˆ์Šต๋‹ˆ๋‹ค.\n- ์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ: systematic_review(1), experimental(5), theoretical(3), survey(1)", "charts": { "year_trend": "์—ฐ๋„๋ณ„ ํŠธ๋ Œ๋“œ ์ฐจํŠธ", "venue_distribution": "์ €๋„ ๋ถ„ํฌ ์ฐจํŠธ", "study_type_distribution": "์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ ์ฐจํŠธ" } }
What are the technical details of RNN?
10
๋‹ค์Œ ์งˆ๋ฌธ์— ๋Œ€ํ•ด ๋…ผ๋ฌธ์„ ๊ฒ€์ƒ‰ํ•˜๊ณ  ๋žญํ‚นํ•˜์—ฌ ๊ฒฐ๊ณผ๋ฅผ ์ œ์‹œํ•˜์„ธ์š”. ์งˆ๋ฌธ: How has deep learning evolved over the past 5 years? ์ถœ๋ ฅ ํ˜•์‹: 1. Evidence Table (Top-K): ๊ฒ€์ƒ‰๋œ ๋…ผ๋ฌธ์˜ ์ƒ์œ„ K๊ฐœ๋ฅผ ํ…Œ์ด๋ธ” ํ˜•์‹์œผ๋กœ ์ œ์‹œ 2. Ranking Breakdown: ๊ฐ ๋…ผ๋ฌธ์˜ ๋žญํ‚น ์ ์ˆ˜ ๊ตฌ์„ฑ ์š”์†Œ๋ฅผ ๊ณต๊ฐœ 3. Answer Summary: ๊ฒ€์ƒ‰ ๊ฒฐ๊ณผ๋ฅผ ์š”์•ฝํ•˜์—ฌ ์ œ์‹œ 4. Charts: ์—ฐ๋„๋ณ„ ํŠธ๋ Œ๋“œ, ์ €๋„ ๋ถ„ํฌ, ์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ ์ฐจํŠธ์— ๋Œ€ํ•œ ์„ค๋ช… ๋ชจ๋“  ์ฃผ์žฅ์€ Evidence Table์˜ ๋…ผ๋ฌธ์— ๊ทผ๊ฑฐํ•ด์•ผ ํ•˜๋ฉฐ, ๋žญํ‚น ๊ทผ๊ฑฐ๋ฅผ ๋ช…์‹œํ•ด์•ผ ํ•ฉ๋‹ˆ๋‹ค.
{ "evidence_table": [ { "rank": 1, "title": "Review of applications of deep learning in veterinary diagnostics and animal health.", "authors": "Xiao Sam, Dhand Navneet K, Wang Zhiyong", "year": 2025, "venue": "Front Vet Sci", "study_type": "systematic_review", "score": 0.651, "url": "https://pubmed.ncbi.nlm.nih.gov/40144529/" }, { "rank": 2, "title": "French guidelines for the diagnosis and management of Tourette syndrome.", "authors": "Hartmann A, Ansquer S, Brefel-Courbon C", "year": 2024, "venue": "Rev Neurol (Paris)", "study_type": "systematic_review", "score": 0.6140000000000001, "url": "https://pubmed.ncbi.nlm.nih.gov/38760282/" }, { "rank": 3, "title": "On the use of deep learning for computer-generated holography.", "authors": "Yu Xuan, Zhang Haomiao, Zhao Zhe", "year": 2025, "venue": "iScience", "study_type": "theoretical", "score": 0.5910000000000001, "url": "https://pubmed.ncbi.nlm.nih.gov/40491959/" }, { "rank": 4, "title": "Machine learning for diabetic foot care: accuracy trends and emerging directions in healthcare AI.", "authors": "Lin Pei-Chun, Li Tsai-Chung, Huang Tzu-Hsuan", "year": 2025, "venue": "Front Public Health", "study_type": "theoretical", "score": 0.5740000000000001, "url": "https://pubmed.ncbi.nlm.nih.gov/40756392/" }, { "rank": 5, "title": "Decoding the Digital Pulse: Bibliometric Analysis of 25 Years in Digital Health Research Through the Journal of Medical Internet Research.", "authors": "Kaczmarczyk Robert, Wilhelm Theresa Isabelle, Roos Jonas", "year": 2024, "venue": "J Med Internet Res", "study_type": "theoretical", "score": 0.554, "url": "https://pubmed.ncbi.nlm.nih.gov/39546778/" }, { "rank": 6, "title": "Mapping the Bibliometrics Landscape of AI in Medicine: Methodological Study.", "authors": "Shi Jin, Bendig David, Vollmar Horst Christian", "year": 2023, "venue": "J Med Internet Res", "study_type": "theoretical", "score": 0.538, "url": "https://pubmed.ncbi.nlm.nih.gov/38064255/" }, { "rank": 7, "title": "Breast Cancer Detection and Diagnosis Using Mammographic Data: Systematic Review.", "authors": "Gardezi Syed Jamal Safdar, Elazab Ahmed, Lei Baiying", "year": 2019, "venue": "J Med Internet Res", "study_type": "systematic_review", "score": 0.534, "url": "https://pubmed.ncbi.nlm.nih.gov/31350843/" }, { "rank": 8, "title": "Hybrid deep learning-based strategy for the hepatocellular carcinoma cancer grade classification of H&E stained liver histopathology images", "authors": "Ajinkya Deshpande, Deep Gupta, Ankit Bhurane", "year": 2024, "venue": "arXiv", "study_type": "theoretical", "score": 0.5305, "url": "https://arxiv.org/abs/2412.03084v2" }, { "rank": 9, "title": "PePR: Performance Per Resource Unit as a Metric to Promote Small-Scale Deep Learning in Medical Image Analysis", "authors": "Raghavendra Selvan, Bob Pepin, Christian Igel", "year": 2024, "venue": "arXiv", "study_type": "theoretical", "score": 0.49849999999999994, "url": "https://arxiv.org/abs/2403.12562v2" }, { "rank": 10, "title": "Inverse Contextual Bandits: Learning How Behavior Evolves over Time", "authors": "Alihan Hรผyรผk, Daniel Jarrett, Mihaela van der Schaar", "year": 2021, "venue": "arXiv", "study_type": "theoretical", "score": 0.49449999999999994, "url": "https://arxiv.org/abs/2107.06317v3" } ], "ranking_breakdown": [ { "paper_id": "40144529", "title": "Review of applications of deep learning in veterinary diagnostics and animal health.", "total_score": 0.651, "query_match": 0.33999999999999997, "recency": 1.0, "study_type_priority": 0.9, "evidence_completeness": 0.8999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "38760282", "title": "French guidelines for the diagnosis and management of Tourette syndrome.", "total_score": 0.6140000000000001, "query_match": 0.26, "recency": 0.9, "study_type_priority": 0.9, "evidence_completeness": 0.9999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "40491959", "title": "On the use of deep learning for computer-generated holography.", "total_score": 0.5910000000000001, "query_match": 0.39, "recency": 1.0, "study_type_priority": 0.5, "evidence_completeness": 0.8999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "40756392", "title": "Machine learning for diabetic foot care: accuracy trends and emerging directions in healthcare AI.", "total_score": 0.5740000000000001, "query_match": 0.31, "recency": 1.0, "study_type_priority": 0.5, "evidence_completeness": 0.9999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "39546778", "title": "Decoding the Digital Pulse: Bibliometric Analysis of 25 Years in Digital Health Research Through the", "total_score": 0.554, "query_match": 0.31, "recency": 0.9, "study_type_priority": 0.5, "evidence_completeness": 0.9999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "38064255", "title": "Mapping the Bibliometrics Landscape of AI in Medicine: Methodological Study.", "total_score": 0.538, "query_match": 0.32, "recency": 0.8, "study_type_priority": 0.5, "evidence_completeness": 0.9999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "31350843", "title": "Breast Cancer Detection and Diagnosis Using Mammographic Data: Systematic Review.", "total_score": 0.534, "query_match": 0.31000000000000005, "recency": 0.4, "study_type_priority": 0.9, "evidence_completeness": 0.9999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "2412.03084", "title": "Hybrid deep learning-based strategy for the hepatocellular carcinoma cancer grade classification of ", "total_score": 0.5305, "query_match": 0.27, "recency": 0.9, "study_type_priority": 0.5, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2403.12562", "title": "PePR: Performance Per Resource Unit as a Metric to Promote Small-Scale Deep Learning in Medical Imag", "total_score": 0.49849999999999994, "query_match": 0.19, "recency": 0.9, "study_type_priority": 0.5, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2107.06317", "title": "Inverse Contextual Bandits: Learning How Behavior Evolves over Time", "total_score": 0.49449999999999994, "query_match": 0.32999999999999996, "recency": 0.6, "study_type_priority": 0.5, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 } ], "answer_summary": "๊ฒ€์ƒ‰ ์งˆ๋ฌธ: How has deep learning evolved over the past 5 years?\n\n๊ฒ€์ƒ‰ ๊ฒฐ๊ณผ ์š”์•ฝ:\n- ์ด 10๊ฐœ์˜ ๋…ผ๋ฌธ์ด ๊ฒ€์ƒ‰๋˜์—ˆ์Šต๋‹ˆ๋‹ค.\n- ์†Œ์Šค๋ณ„ ๋ถ„ํฌ: pubmed(7), arxiv(3)\n- ์—ฐ๋„ ๋ฒ”์œ„: 2019-2025\n- ํ‰๊ท  ๋žญํ‚น ์ ์ˆ˜: 0.558\n\n์ฃผ์š” ๋ฐœ๊ฒฌ:\n- ์ƒ์œ„ ๋…ผ๋ฌธ๋“ค์€ ์ฃผ๋กœ Front Vet Sci, Rev Neurol (Paris), iScience์—์„œ ๋ฐœํ‘œ๋˜์—ˆ์Šต๋‹ˆ๋‹ค.\n- ์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ: systematic_review(3), theoretical(7)", "charts": { "year_trend": "์—ฐ๋„๋ณ„ ํŠธ๋ Œ๋“œ ์ฐจํŠธ", "venue_distribution": "์ €๋„ ๋ถ„ํฌ ์ฐจํŠธ", "study_type_distribution": "์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ ์ฐจํŠธ" } }
How has deep learning evolved over the past 5 years?
10
๋‹ค์Œ ์งˆ๋ฌธ์— ๋Œ€ํ•ด ๋…ผ๋ฌธ์„ ๊ฒ€์ƒ‰ํ•˜๊ณ  ๋žญํ‚นํ•˜์—ฌ ๊ฒฐ๊ณผ๋ฅผ ์ œ์‹œํ•˜์„ธ์š”. ์งˆ๋ฌธ: Explain the architecture of ResNet ์ถœ๋ ฅ ํ˜•์‹: 1. Evidence Table (Top-K): ๊ฒ€์ƒ‰๋œ ๋…ผ๋ฌธ์˜ ์ƒ์œ„ K๊ฐœ๋ฅผ ํ…Œ์ด๋ธ” ํ˜•์‹์œผ๋กœ ์ œ์‹œ 2. Ranking Breakdown: ๊ฐ ๋…ผ๋ฌธ์˜ ๋žญํ‚น ์ ์ˆ˜ ๊ตฌ์„ฑ ์š”์†Œ๋ฅผ ๊ณต๊ฐœ 3. Answer Summary: ๊ฒ€์ƒ‰ ๊ฒฐ๊ณผ๋ฅผ ์š”์•ฝํ•˜์—ฌ ์ œ์‹œ 4. Charts: ์—ฐ๋„๋ณ„ ํŠธ๋ Œ๋“œ, ์ €๋„ ๋ถ„ํฌ, ์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ ์ฐจํŠธ์— ๋Œ€ํ•œ ์„ค๋ช… ๋ชจ๋“  ์ฃผ์žฅ์€ Evidence Table์˜ ๋…ผ๋ฌธ์— ๊ทผ๊ฑฐํ•ด์•ผ ํ•˜๋ฉฐ, ๋žญํ‚น ๊ทผ๊ฑฐ๋ฅผ ๋ช…์‹œํ•ด์•ผ ํ•ฉ๋‹ˆ๋‹ค.
{ "evidence_table": [ { "rank": 1, "title": "Artificial Intelligence-Based Predictive Modeling for Early Detection of Sepsis in Hospitalized Patients: A Systematic Review and Meta-Analysis.", "authors": "Abbas Ghulam Husain, Sen Palash, Giri Oviya Anjali", "year": 2025, "venue": "Crit Care Explor", "study_type": "meta_analysis", "score": 0.638, "url": "https://pubmed.ncbi.nlm.nih.gov/41348160/" }, { "rank": 2, "title": "Systematic Review of Artificial Intelligence and Electrocardiography for Cardiovascular Disease Diagnosis.", "authors": "Velandia Hernando, Pardo Aldo, Vera Marรญa Isabel", "year": 2025, "venue": "Bioengineering (Basel)", "study_type": "systematic_review", "score": 0.6270000000000001, "url": "https://pubmed.ncbi.nlm.nih.gov/41301204/" }, { "rank": 3, "title": "Explainable Convolutional Neural Networks for the identification of the Ampullariidae genus.", "authors": "Duwa Rabi Suraj, Suraj Kabir Salihu", "year": 2026, "venue": "Parasitol Int", "study_type": "theoretical", "score": 0.6220000000000001, "url": "https://pubmed.ncbi.nlm.nih.gov/41271157/" }, { "rank": 4, "title": "Towards Assessing Spread in Sets of Software Architecture Designs", "authors": "Vittorio Cortellessa, J. Andres Diaz-Pace, Daniele Di Pompeo", "year": 2024, "venue": "arXiv", "study_type": "experimental", "score": 0.6145, "url": "https://arxiv.org/abs/2402.19171v1" }, { "rank": 5, "title": "Hybrid AI Pipeline for Laboratory Detection of Internal Potato Defects Using 2D RGB Imaging.", "authors": "Hamdi Slim, Loukil Kais, Boubaker Adem Haj", "year": 2025, "venue": "J Imaging", "study_type": "experimental", "score": 0.611, "url": "https://pubmed.ncbi.nlm.nih.gov/41440571/" }, { "rank": 6, "title": "Quotient Network -- A Network Similar to ResNet but Learning Quotients", "authors": "Peng Hui, Jiamuyang Zhao, Changxin Li", "year": 2025, "venue": "arXiv", "study_type": "experimental", "score": 0.5945, "url": "https://arxiv.org/abs/2506.00992v1" }, { "rank": 7, "title": "Boosting brain tumor detection with an optimized ResNet and explainability via Grad-CAM and LIME.", "authors": "Afnaan K, Arunbalaji C G, Singh Tripty", "year": 2025, "venue": "Brain Inform", "study_type": "experimental", "score": 0.5870000000000001, "url": "https://pubmed.ncbi.nlm.nih.gov/41348154/" }, { "rank": 8, "title": "A comparative evaluation of explainability techniques for image data.", "authors": "Skliarov Mykyta, Shawi Radwa El, Dhaoui Chedia", "year": 2025, "venue": "Sci Rep", "study_type": "experimental", "score": 0.5870000000000001, "url": "https://pubmed.ncbi.nlm.nih.gov/41290895/" }, { "rank": 9, "title": "Block-Level Genetic Algorithm Optimization for ResNet Endoscopic Image Classification.", "authors": "De Souza Junior Gilberto R, De Sousa Filho Gilberto F, Cabral Lucidio Dos Anjos F", "year": 2025, "venue": "Annu Int Conf IEEE Eng Med Biol Soc", "study_type": "theoretical", "score": 0.5859999999999999, "url": "https://pubmed.ncbi.nlm.nih.gov/41337417/" }, { "rank": 10, "title": "Multi-phase deep learning framework with Multiscale Adaptive Swin Transformer and embedding attention for precision lung nodule detection and classification.", "authors": "M Dhayalini, B Revathi Alias Ponmozhi", "year": 2025, "venue": "Sci Rep", "study_type": "experimental", "score": 0.5710000000000001, "url": "https://pubmed.ncbi.nlm.nih.gov/41390502/" } ], "ranking_breakdown": [ { "paper_id": "41348160", "title": "Artificial Intelligence-Based Predictive Modeling for Early Detection of Sepsis in Hospitalized Pati", "total_score": 0.638, "query_match": 0.22, "recency": 1.0, "study_type_priority": 1.0, "evidence_completeness": 0.9999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "41301204", "title": "Systematic Review of Artificial Intelligence and Electrocardiography for Cardiovascular Disease Diag", "total_score": 0.6270000000000001, "query_match": 0.28, "recency": 1.0, "study_type_priority": 0.9, "evidence_completeness": 0.8999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "41271157", "title": "Explainable Convolutional Neural Networks for the identification of the Ampullariidae genus.", "total_score": 0.6220000000000001, "query_match": 0.38, "recency": 1.1, "study_type_priority": 0.5, "evidence_completeness": 0.9999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "2402.19171", "title": "Towards Assessing Spread in Sets of Software Architecture Designs", "total_score": 0.6145, "query_match": 0.38, "recency": 0.9, "study_type_priority": 0.7, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "41440571", "title": "Hybrid AI Pipeline for Laboratory Detection of Internal Potato Defects Using 2D RGB Imaging.", "total_score": 0.611, "query_match": 0.33999999999999997, "recency": 1.0, "study_type_priority": 0.7, "evidence_completeness": 0.8999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "2506.00992", "title": "Quotient Network -- A Network Similar to ResNet but Learning Quotients", "total_score": 0.5945, "query_match": 0.28, "recency": 1.0, "study_type_priority": 0.7, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "41348154", "title": "Boosting brain tumor detection with an optimized ResNet and explainability via Grad-CAM and LIME.", "total_score": 0.5870000000000001, "query_match": 0.28, "recency": 1.0, "study_type_priority": 0.7, "evidence_completeness": 0.8999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "41290895", "title": "A comparative evaluation of explainability techniques for image data.", "total_score": 0.5870000000000001, "query_match": 0.28, "recency": 1.0, "study_type_priority": 0.7, "evidence_completeness": 0.8999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "41337417", "title": "Block-Level Genetic Algorithm Optimization for ResNet Endoscopic Image Classification.", "total_score": 0.5859999999999999, "query_match": 0.33999999999999997, "recency": 1.0, "study_type_priority": 0.5, "evidence_completeness": 0.9999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "41390502", "title": "Multi-phase deep learning framework with Multiscale Adaptive Swin Transformer and embedding attentio", "total_score": 0.5710000000000001, "query_match": 0.24, "recency": 1.0, "study_type_priority": 0.7, "evidence_completeness": 0.8999999999999999, "duplicate_penalty": 0.0 } ], "answer_summary": "๊ฒ€์ƒ‰ ์งˆ๋ฌธ: Explain the architecture of ResNet\n\n๊ฒ€์ƒ‰ ๊ฒฐ๊ณผ ์š”์•ฝ:\n- ์ด 10๊ฐœ์˜ ๋…ผ๋ฌธ์ด ๊ฒ€์ƒ‰๋˜์—ˆ์Šต๋‹ˆ๋‹ค.\n- ์†Œ์Šค๋ณ„ ๋ถ„ํฌ: pubmed(8), arxiv(2)\n- ์—ฐ๋„ ๋ฒ”์œ„: 2024-2026\n- ํ‰๊ท  ๋žญํ‚น ์ ์ˆ˜: 0.604\n\n์ฃผ์š” ๋ฐœ๊ฒฌ:\n- ์ƒ์œ„ ๋…ผ๋ฌธ๋“ค์€ ์ฃผ๋กœ Crit Care Explor, Bioengineering (Basel), Parasitol Int์—์„œ ๋ฐœํ‘œ๋˜์—ˆ์Šต๋‹ˆ๋‹ค.\n- ์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ: meta_analysis(1), systematic_review(1), theoretical(2), experimental(6)", "charts": { "year_trend": "์—ฐ๋„๋ณ„ ํŠธ๋ Œ๋“œ ์ฐจํŠธ", "venue_distribution": "์ €๋„ ๋ถ„ํฌ ์ฐจํŠธ", "study_type_distribution": "์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ ์ฐจํŠธ" } }
Explain the architecture of ResNet
10
๋‹ค์Œ ์งˆ๋ฌธ์— ๋Œ€ํ•ด ๋…ผ๋ฌธ์„ ๊ฒ€์ƒ‰ํ•˜๊ณ  ๋žญํ‚นํ•˜์—ฌ ๊ฒฐ๊ณผ๋ฅผ ์ œ์‹œํ•˜์„ธ์š”. ์งˆ๋ฌธ: What are the technical details of RNN? ์ถœ๋ ฅ ํ˜•์‹: 1. Evidence Table (Top-K): ๊ฒ€์ƒ‰๋œ ๋…ผ๋ฌธ์˜ ์ƒ์œ„ K๊ฐœ๋ฅผ ํ…Œ์ด๋ธ” ํ˜•์‹์œผ๋กœ ์ œ์‹œ 2. Ranking Breakdown: ๊ฐ ๋…ผ๋ฌธ์˜ ๋žญํ‚น ์ ์ˆ˜ ๊ตฌ์„ฑ ์š”์†Œ๋ฅผ ๊ณต๊ฐœ 3. Answer Summary: ๊ฒ€์ƒ‰ ๊ฒฐ๊ณผ๋ฅผ ์š”์•ฝํ•˜์—ฌ ์ œ์‹œ 4. Charts: ์—ฐ๋„๋ณ„ ํŠธ๋ Œ๋“œ, ์ €๋„ ๋ถ„ํฌ, ์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ ์ฐจํŠธ์— ๋Œ€ํ•œ ์„ค๋ช… ๋ชจ๋“  ์ฃผ์žฅ์€ Evidence Table์˜ ๋…ผ๋ฌธ์— ๊ทผ๊ฑฐํ•ด์•ผ ํ•˜๋ฉฐ, ๋žญํ‚น ๊ทผ๊ฑฐ๋ฅผ ๋ช…์‹œํ•ด์•ผ ํ•ฉ๋‹ˆ๋‹ค.
{ "evidence_table": [ { "rank": 1, "title": "Deep learning in surgical process modeling: A systematic review of workflow recognition.", "authors": "Liu Zhenzhong, Chen Kelong, Wang Shuai", "year": 2025, "venue": "J Biomed Inform", "study_type": "systematic_review", "score": 0.6271428571428572, "url": "https://pubmed.ncbi.nlm.nih.gov/39832608/" }, { "rank": 2, "title": "A hybrid deep learning framework for early detection of diabetic retinopathy using retinal fundus images.", "authors": "Sushith Mishmala, Sathiya A, Kalaipoonguzhali V", "year": 2025, "venue": "Sci Rep", "study_type": "experimental", "score": 0.6042857142857143, "url": "https://pubmed.ncbi.nlm.nih.gov/40307328/" }, { "rank": 3, "title": "Overcoming the Barrier of Incompleteness: A Hyperspectral Image Classification Full Model.", "authors": "Yang Jiaqi, Du Bo, Zhang Liangpei", "year": 2024, "venue": "IEEE Trans Neural Netw Learn Syst", "study_type": "experimental", "score": 0.5978571428571429, "url": "https://pubmed.ncbi.nlm.nih.gov/37279129/" }, { "rank": 4, "title": "ElectroPhysiomeGAN: Generation of Biophysical Neuron Model Parameters from Recorded Electrophysiological Responses.", "authors": "Kim Jimin, Peng Minxian, Chen Shuqi", "year": 2025, "venue": "bioRxiv", "study_type": "experimental", "score": 0.5892857142857143, "url": "https://pubmed.ncbi.nlm.nih.gov/40501917/" }, { "rank": 5, "title": "Precision diagnosis of burn injuries using imaging and predictive modeling for clinical applications.", "authors": "Rangaiah Pramod K B, Kumar B P Pradeep, Huss Fredrik", "year": 2025, "venue": "Sci Rep", "study_type": "experimental", "score": 0.5871428571428572, "url": "https://pubmed.ncbi.nlm.nih.gov/40038450/" }, { "rank": 6, "title": "Integration of RNN and CatBoost models in a tea-waste biochar filtration system for toxic organic pollutant removal efficiency prediction.", "authors": "Jha Stuti, Gaur Rama, Shahabuddin Syed", "year": 2025, "venue": "RSC Adv", "study_type": "theoretical", "score": 0.5607142857142857, "url": "https://pubmed.ncbi.nlm.nih.gov/40746784/" }, { "rank": 7, "title": "A recurrent neural network for soft sensor development using CHO stable pools in fed-batch process for SARS-CoV-2 spike protein production as a vaccine antigen.", "authors": "Reyes Sebastian-Juan, Voyer Robert, Durocher Yves", "year": 2025, "venue": "Biotechnol Prog", "study_type": "experimental", "score": 0.5585714285714285, "url": "https://pubmed.ncbi.nlm.nih.gov/40454837/" }, { "rank": 8, "title": "What's behind tight deadlines? Business causes of technical debt", "authors": "Rodrigo Rebouรงas de Almeida, Christoph Treude, Uirรก Kulesza", "year": 2021, "venue": "arXiv", "study_type": "survey", "score": 0.5539285714285714, "url": "https://arxiv.org/abs/2104.09330v5" }, { "rank": 9, "title": "Proceedings to the 27th Workshop \"What Comes Beyond the Standard Models\" Bled, July 8-17, 2024", "authors": "R. Bernabei, P. Belli, A. Bussolotti", "year": 2025, "venue": "arXiv", "study_type": "theoretical", "score": 0.5510714285714285, "url": "https://arxiv.org/abs/2504.17803v1" }, { "rank": 10, "title": "A One-Dimensional Energy Balance Model Parameterization for the Formation of CO2 Ice on the Surfaces of Eccentric Extrasolar Planets", "authors": "Vidya Venkatesan, Aomawa L. Shields, Russell Deitrick", "year": 2025, "venue": "arXiv", "study_type": "theoretical", "score": 0.5510714285714285, "url": "https://arxiv.org/abs/2501.11667v1" } ], "ranking_breakdown": [ { "paper_id": "39832608", "title": "Deep learning in surgical process modeling: A systematic review of workflow recognition.", "total_score": 0.6271428571428572, "query_match": 0.24285714285714283, "recency": 1.0, "study_type_priority": 0.9, "evidence_completeness": 0.9999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "40307328", "title": "A hybrid deep learning framework for early detection of diabetic retinopathy using retinal fundus im", "total_score": 0.6042857142857143, "query_match": 0.2857142857142857, "recency": 1.0, "study_type_priority": 0.7, "evidence_completeness": 0.9999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "37279129", "title": "Overcoming the Barrier of Incompleteness: A Hyperspectral Image Classification Full Model.", "total_score": 0.5978571428571429, "query_match": 0.3571428571428571, "recency": 0.9, "study_type_priority": 0.7, "evidence_completeness": 0.8999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "40501917", "title": "ElectroPhysiomeGAN: Generation of Biophysical Neuron Model Parameters from Recorded Electrophysiolog", "total_score": 0.5892857142857143, "query_match": 0.2857142857142857, "recency": 1.0, "study_type_priority": 0.7, "evidence_completeness": 0.8999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "40038450", "title": "Precision diagnosis of burn injuries using imaging and predictive modeling for clinical applications", "total_score": 0.5871428571428572, "query_match": 0.24285714285714283, "recency": 1.0, "study_type_priority": 0.7, "evidence_completeness": 0.9999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "40746784", "title": "Integration of RNN and CatBoost models in a tea-waste biochar filtration system for toxic organic po", "total_score": 0.5607142857142857, "query_match": 0.3142857142857143, "recency": 1.0, "study_type_priority": 0.5, "evidence_completeness": 0.8999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "40454837", "title": "A recurrent neural network for soft sensor development using CHO stable pools in fed-batch process f", "total_score": 0.5585714285714285, "query_match": 0.1714285714285714, "recency": 1.0, "study_type_priority": 0.7, "evidence_completeness": 0.9999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "2104.09330", "title": "What's behind tight deadlines? Business causes of technical debt", "total_score": 0.5539285714285714, "query_match": 0.42857142857142855, "recency": 0.6, "study_type_priority": 0.6, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2504.17803", "title": "Proceedings to the 27th Workshop \"What Comes Beyond the Standard Models\" Bled, July 8-17, 2024", "total_score": 0.5510714285714285, "query_match": 0.2714285714285714, "recency": 1.0, "study_type_priority": 0.5, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2501.11667", "title": "A One-Dimensional Energy Balance Model Parameterization for the Formation of CO2 Ice on the Surfaces", "total_score": 0.5510714285714285, "query_match": 0.2714285714285714, "recency": 1.0, "study_type_priority": 0.5, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 } ], "answer_summary": "๊ฒ€์ƒ‰ ์งˆ๋ฌธ: What are the technical details of RNN?\n\n๊ฒ€์ƒ‰ ๊ฒฐ๊ณผ ์š”์•ฝ:\n- ์ด 10๊ฐœ์˜ ๋…ผ๋ฌธ์ด ๊ฒ€์ƒ‰๋˜์—ˆ์Šต๋‹ˆ๋‹ค.\n- ์†Œ์Šค๋ณ„ ๋ถ„ํฌ: pubmed(7), arxiv(3)\n- ์—ฐ๋„ ๋ฒ”์œ„: 2021-2025\n- ํ‰๊ท  ๋žญํ‚น ์ ์ˆ˜: 0.578\n\n์ฃผ์š” ๋ฐœ๊ฒฌ:\n- ์ƒ์œ„ ๋…ผ๋ฌธ๋“ค์€ ์ฃผ๋กœ J Biomed Inform, Sci Rep, IEEE Trans Neural Netw Learn Syst์—์„œ ๋ฐœํ‘œ๋˜์—ˆ์Šต๋‹ˆ๋‹ค.\n- ์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ: systematic_review(1), experimental(5), theoretical(3), survey(1)", "charts": { "year_trend": "์—ฐ๋„๋ณ„ ํŠธ๋ Œ๋“œ ์ฐจํŠธ", "venue_distribution": "์ €๋„ ๋ถ„ํฌ ์ฐจํŠธ", "study_type_distribution": "์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ ์ฐจํŠธ" } }
What are the technical details of RNN?
10
๋‹ค์Œ ์งˆ๋ฌธ์— ๋Œ€ํ•ด ๋…ผ๋ฌธ์„ ๊ฒ€์ƒ‰ํ•˜๊ณ  ๋žญํ‚นํ•˜์—ฌ ๊ฒฐ๊ณผ๋ฅผ ์ œ์‹œํ•˜์„ธ์š”. ์งˆ๋ฌธ: What do systematic reviews say about large language models? ์ถœ๋ ฅ ํ˜•์‹: 1. Evidence Table (Top-K): ๊ฒ€์ƒ‰๋œ ๋…ผ๋ฌธ์˜ ์ƒ์œ„ K๊ฐœ๋ฅผ ํ…Œ์ด๋ธ” ํ˜•์‹์œผ๋กœ ์ œ์‹œ 2. Ranking Breakdown: ๊ฐ ๋…ผ๋ฌธ์˜ ๋žญํ‚น ์ ์ˆ˜ ๊ตฌ์„ฑ ์š”์†Œ๋ฅผ ๊ณต๊ฐœ 3. Answer Summary: ๊ฒ€์ƒ‰ ๊ฒฐ๊ณผ๋ฅผ ์š”์•ฝํ•˜์—ฌ ์ œ์‹œ 4. Charts: ์—ฐ๋„๋ณ„ ํŠธ๋ Œ๋“œ, ์ €๋„ ๋ถ„ํฌ, ์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ ์ฐจํŠธ์— ๋Œ€ํ•œ ์„ค๋ช… ๋ชจ๋“  ์ฃผ์žฅ์€ Evidence Table์˜ ๋…ผ๋ฌธ์— ๊ทผ๊ฑฐํ•ด์•ผ ํ•˜๋ฉฐ, ๋žญํ‚น ๊ทผ๊ฑฐ๋ฅผ ๋ช…์‹œํ•ด์•ผ ํ•ฉ๋‹ˆ๋‹ค.
{ "evidence_table": [ { "rank": 1, "title": "Enhancing Systematic Reviews with Large Language Models: Using GPT-4 and Kimi", "authors": "Dandan Chen Kaptur, Yue Huang, Xuejun Ryan Ji", "year": 2025, "venue": "arXiv", "study_type": "systematic_review", "score": 0.6852777777777779, "url": "https://arxiv.org/abs/2504.20276v1" }, { "rank": 2, "title": "Efficacy of Large Language Models in Systematic Reviews", "authors": "Aaditya Shah, Shridhar Mehendale, Siddha Kanthi", "year": 2024, "venue": "arXiv", "study_type": "systematic_review", "score": 0.6669444444444445, "url": "https://arxiv.org/abs/2408.04646v2" }, { "rank": 3, "title": "A systematic review of research on large language models for computer programming education", "authors": "Meina Zhu, Lanyu Xu, Barbara Ericson", "year": 2025, "venue": "arXiv", "study_type": "systematic_review", "score": 0.6647222222222222, "url": "https://arxiv.org/abs/2506.21818v1" }, { "rank": 4, "title": "Exploring the use of a Large Language Model for data extraction in systematic reviews: a rapid feasibility study", "authors": "Lena Schmidt, Kaitlyn Hair, Sergio Graziosi", "year": 2024, "venue": "arXiv", "study_type": "systematic_review", "score": 0.6580555555555556, "url": "https://arxiv.org/abs/2405.14445v2" }, { "rank": 5, "title": "Advances in Pre-trained Language Models for Domain-Specific Text Classification: A Systematic Review", "authors": "Zhyar Rzgar K. Rostam, Gรกbor Kertรฉsz", "year": 2025, "venue": "arXiv", "study_type": "systematic_review", "score": 0.6558333333333334, "url": "https://arxiv.org/abs/2510.17892v1" }, { "rank": 6, "title": "What Do We Mean When We Talk about Trust in Social Media? A Systematic Review", "authors": "Yixuan Zhang, Joseph D Gaggiano, Nutchanon Yongsatianchot", "year": 2023, "venue": "arXiv", "study_type": "systematic_review", "score": 0.6247222222222223, "url": "https://arxiv.org/abs/2302.03671v1" }, { "rank": 7, "title": "Systematic Outliers in Large Language Models", "authors": "Yongqi An, Xu Zhao, Tao Yu", "year": 2025, "venue": "arXiv", "study_type": "experimental", "score": 0.6247222222222222, "url": "https://arxiv.org/abs/2502.06415v2" }, { "rank": 8, "title": "Game Theory Meets Large Language Models: A Systematic Survey with Taxonomy and New Frontiers", "authors": "Haoran Sun, Yusen Wu, Peng Wang", "year": 2025, "venue": "arXiv", "study_type": "survey", "score": 0.5913888888888889, "url": "https://arxiv.org/abs/2502.09053v2" }, { "rank": 9, "title": "Effectiveness of Chain-of-Thought in Distilling Reasoning Capability from Large Language Models", "authors": "Cong-Thanh Do, Rama Doddipatla, Kate Knill", "year": 2025, "venue": "arXiv", "study_type": "experimental", "score": 0.5891666666666666, "url": "https://arxiv.org/abs/2511.05184v1" }, { "rank": 10, "title": "Soft Inductive Bias Approach via Explicit Reasoning Perspectives in Inappropriate Utterance Detection Using Large Language Models", "authors": "Ju-Young Kim, Ji-Hong Park, Se-Yeon Lee", "year": 2025, "venue": "arXiv", "study_type": "experimental", "score": 0.5891666666666666, "url": "https://arxiv.org/abs/2512.08480v1" } ], "ranking_breakdown": [ { "paper_id": "2504.20276", "title": "Enhancing Systematic Reviews with Large Language Models: Using GPT-4 and Kimi", "total_score": 0.6852777777777779, "query_match": 0.4444444444444444, "recency": 1.0, "study_type_priority": 0.9, "evidence_completeness": 0.85, "duplicate_penalty": 0.0 }, { "paper_id": "2408.04646", "title": "Efficacy of Large Language Models in Systematic Reviews", "total_score": 0.6669444444444445, "query_match": 0.4111111111111111, "recency": 0.9, "study_type_priority": 0.9, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2506.21818", "title": "A systematic review of research on large language models for computer programming education", "total_score": 0.6647222222222222, "query_match": 0.3555555555555555, "recency": 1.0, "study_type_priority": 0.9, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2405.14445", "title": "Exploring the use of a Large Language Model for data extraction in systematic reviews: a rapid feasi", "total_score": 0.6580555555555556, "query_match": 0.38888888888888884, "recency": 0.9, "study_type_priority": 0.9, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2510.17892", "title": "Advances in Pre-trained Language Models for Domain-Specific Text Classification: A Systematic Review", "total_score": 0.6558333333333334, "query_match": 0.3333333333333333, "recency": 1.0, "study_type_priority": 0.9, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2302.03671", "title": "What Do We Mean When We Talk about Trust in Social Media? A Systematic Review", "total_score": 0.6247222222222223, "query_match": 0.3555555555555555, "recency": 0.8, "study_type_priority": 0.9, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2502.06415", "title": "Systematic Outliers in Large Language Models", "total_score": 0.6247222222222222, "query_match": 0.3555555555555555, "recency": 1.0, "study_type_priority": 0.7, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2502.09053", "title": "Game Theory Meets Large Language Models: A Systematic Survey with Taxonomy and New Frontiers", "total_score": 0.5913888888888889, "query_match": 0.3222222222222222, "recency": 1.0, "study_type_priority": 0.6, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2511.05184", "title": "Effectiveness of Chain-of-Thought in Distilling Reasoning Capability from Large Language Models", "total_score": 0.5891666666666666, "query_match": 0.26666666666666666, "recency": 1.0, "study_type_priority": 0.7, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2512.08480", "title": "Soft Inductive Bias Approach via Explicit Reasoning Perspectives in Inappropriate Utterance Detectio", "total_score": 0.5891666666666666, "query_match": 0.26666666666666666, "recency": 1.0, "study_type_priority": 0.7, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 } ], "answer_summary": "๊ฒ€์ƒ‰ ์งˆ๋ฌธ: What do systematic reviews say about large language models?\n\n๊ฒ€์ƒ‰ ๊ฒฐ๊ณผ ์š”์•ฝ:\n- ์ด 10๊ฐœ์˜ ๋…ผ๋ฌธ์ด ๊ฒ€์ƒ‰๋˜์—ˆ์Šต๋‹ˆ๋‹ค.\n- ์†Œ์Šค๋ณ„ ๋ถ„ํฌ: arxiv(10)\n- ์—ฐ๋„ ๋ฒ”์œ„: 2023-2025\n- ํ‰๊ท  ๋žญํ‚น ์ ์ˆ˜: 0.635\n\n์ฃผ์š” ๋ฐœ๊ฒฌ:\n- ์ƒ์œ„ ๋…ผ๋ฌธ๋“ค์€ ์ฃผ๋กœ arXiv, arXiv, arXiv์—์„œ ๋ฐœํ‘œ๋˜์—ˆ์Šต๋‹ˆ๋‹ค.\n- ์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ: systematic_review(6), experimental(3), survey(1)", "charts": { "year_trend": "์—ฐ๋„๋ณ„ ํŠธ๋ Œ๋“œ ์ฐจํŠธ", "venue_distribution": "์ €๋„ ๋ถ„ํฌ ์ฐจํŠธ", "study_type_distribution": "์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ ์ฐจํŠธ" } }
What do systematic reviews say about large language models?
10
๋‹ค์Œ ์งˆ๋ฌธ์— ๋Œ€ํ•ด ๋…ผ๋ฌธ์„ ๊ฒ€์ƒ‰ํ•˜๊ณ  ๋žญํ‚นํ•˜์—ฌ ๊ฒฐ๊ณผ๋ฅผ ์ œ์‹œํ•˜์„ธ์š”. ์งˆ๋ฌธ: What are the technical details of RNN? ์ถœ๋ ฅ ํ˜•์‹: 1. Evidence Table (Top-K): ๊ฒ€์ƒ‰๋œ ๋…ผ๋ฌธ์˜ ์ƒ์œ„ K๊ฐœ๋ฅผ ํ…Œ์ด๋ธ” ํ˜•์‹์œผ๋กœ ์ œ์‹œ 2. Ranking Breakdown: ๊ฐ ๋…ผ๋ฌธ์˜ ๋žญํ‚น ์ ์ˆ˜ ๊ตฌ์„ฑ ์š”์†Œ๋ฅผ ๊ณต๊ฐœ 3. Answer Summary: ๊ฒ€์ƒ‰ ๊ฒฐ๊ณผ๋ฅผ ์š”์•ฝํ•˜์—ฌ ์ œ์‹œ 4. Charts: ์—ฐ๋„๋ณ„ ํŠธ๋ Œ๋“œ, ์ €๋„ ๋ถ„ํฌ, ์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ ์ฐจํŠธ์— ๋Œ€ํ•œ ์„ค๋ช… ๋ชจ๋“  ์ฃผ์žฅ์€ Evidence Table์˜ ๋…ผ๋ฌธ์— ๊ทผ๊ฑฐํ•ด์•ผ ํ•˜๋ฉฐ, ๋žญํ‚น ๊ทผ๊ฑฐ๋ฅผ ๋ช…์‹œํ•ด์•ผ ํ•ฉ๋‹ˆ๋‹ค.
{ "evidence_table": [ { "rank": 1, "title": "Deep learning in surgical process modeling: A systematic review of workflow recognition.", "authors": "Liu Zhenzhong, Chen Kelong, Wang Shuai", "year": 2025, "venue": "J Biomed Inform", "study_type": "systematic_review", "score": 0.6271428571428572, "url": "https://pubmed.ncbi.nlm.nih.gov/39832608/" }, { "rank": 2, "title": "A hybrid deep learning framework for early detection of diabetic retinopathy using retinal fundus images.", "authors": "Sushith Mishmala, Sathiya A, Kalaipoonguzhali V", "year": 2025, "venue": "Sci Rep", "study_type": "experimental", "score": 0.6042857142857143, "url": "https://pubmed.ncbi.nlm.nih.gov/40307328/" }, { "rank": 3, "title": "Overcoming the Barrier of Incompleteness: A Hyperspectral Image Classification Full Model.", "authors": "Yang Jiaqi, Du Bo, Zhang Liangpei", "year": 2024, "venue": "IEEE Trans Neural Netw Learn Syst", "study_type": "experimental", "score": 0.5978571428571429, "url": "https://pubmed.ncbi.nlm.nih.gov/37279129/" }, { "rank": 4, "title": "ElectroPhysiomeGAN: Generation of Biophysical Neuron Model Parameters from Recorded Electrophysiological Responses.", "authors": "Kim Jimin, Peng Minxian, Chen Shuqi", "year": 2025, "venue": "bioRxiv", "study_type": "experimental", "score": 0.5892857142857143, "url": "https://pubmed.ncbi.nlm.nih.gov/40501917/" }, { "rank": 5, "title": "Precision diagnosis of burn injuries using imaging and predictive modeling for clinical applications.", "authors": "Rangaiah Pramod K B, Kumar B P Pradeep, Huss Fredrik", "year": 2025, "venue": "Sci Rep", "study_type": "experimental", "score": 0.5871428571428572, "url": "https://pubmed.ncbi.nlm.nih.gov/40038450/" }, { "rank": 6, "title": "Integration of RNN and CatBoost models in a tea-waste biochar filtration system for toxic organic pollutant removal efficiency prediction.", "authors": "Jha Stuti, Gaur Rama, Shahabuddin Syed", "year": 2025, "venue": "RSC Adv", "study_type": "theoretical", "score": 0.5607142857142857, "url": "https://pubmed.ncbi.nlm.nih.gov/40746784/" }, { "rank": 7, "title": "A recurrent neural network for soft sensor development using CHO stable pools in fed-batch process for SARS-CoV-2 spike protein production as a vaccine antigen.", "authors": "Reyes Sebastian-Juan, Voyer Robert, Durocher Yves", "year": 2025, "venue": "Biotechnol Prog", "study_type": "experimental", "score": 0.5585714285714285, "url": "https://pubmed.ncbi.nlm.nih.gov/40454837/" }, { "rank": 8, "title": "What's behind tight deadlines? Business causes of technical debt", "authors": "Rodrigo Rebouรงas de Almeida, Christoph Treude, Uirรก Kulesza", "year": 2021, "venue": "arXiv", "study_type": "survey", "score": 0.5539285714285714, "url": "https://arxiv.org/abs/2104.09330v5" }, { "rank": 9, "title": "Proceedings to the 27th Workshop \"What Comes Beyond the Standard Models\" Bled, July 8-17, 2024", "authors": "R. Bernabei, P. Belli, A. Bussolotti", "year": 2025, "venue": "arXiv", "study_type": "theoretical", "score": 0.5510714285714285, "url": "https://arxiv.org/abs/2504.17803v1" }, { "rank": 10, "title": "A One-Dimensional Energy Balance Model Parameterization for the Formation of CO2 Ice on the Surfaces of Eccentric Extrasolar Planets", "authors": "Vidya Venkatesan, Aomawa L. Shields, Russell Deitrick", "year": 2025, "venue": "arXiv", "study_type": "theoretical", "score": 0.5510714285714285, "url": "https://arxiv.org/abs/2501.11667v1" } ], "ranking_breakdown": [ { "paper_id": "39832608", "title": "Deep learning in surgical process modeling: A systematic review of workflow recognition.", "total_score": 0.6271428571428572, "query_match": 0.24285714285714283, "recency": 1.0, "study_type_priority": 0.9, "evidence_completeness": 0.9999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "40307328", "title": "A hybrid deep learning framework for early detection of diabetic retinopathy using retinal fundus im", "total_score": 0.6042857142857143, "query_match": 0.2857142857142857, "recency": 1.0, "study_type_priority": 0.7, "evidence_completeness": 0.9999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "37279129", "title": "Overcoming the Barrier of Incompleteness: A Hyperspectral Image Classification Full Model.", "total_score": 0.5978571428571429, "query_match": 0.3571428571428571, "recency": 0.9, "study_type_priority": 0.7, "evidence_completeness": 0.8999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "40501917", "title": "ElectroPhysiomeGAN: Generation of Biophysical Neuron Model Parameters from Recorded Electrophysiolog", "total_score": 0.5892857142857143, "query_match": 0.2857142857142857, "recency": 1.0, "study_type_priority": 0.7, "evidence_completeness": 0.8999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "40038450", "title": "Precision diagnosis of burn injuries using imaging and predictive modeling for clinical applications", "total_score": 0.5871428571428572, "query_match": 0.24285714285714283, "recency": 1.0, "study_type_priority": 0.7, "evidence_completeness": 0.9999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "40746784", "title": "Integration of RNN and CatBoost models in a tea-waste biochar filtration system for toxic organic po", "total_score": 0.5607142857142857, "query_match": 0.3142857142857143, "recency": 1.0, "study_type_priority": 0.5, "evidence_completeness": 0.8999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "40454837", "title": "A recurrent neural network for soft sensor development using CHO stable pools in fed-batch process f", "total_score": 0.5585714285714285, "query_match": 0.1714285714285714, "recency": 1.0, "study_type_priority": 0.7, "evidence_completeness": 0.9999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "2104.09330", "title": "What's behind tight deadlines? Business causes of technical debt", "total_score": 0.5539285714285714, "query_match": 0.42857142857142855, "recency": 0.6, "study_type_priority": 0.6, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2504.17803", "title": "Proceedings to the 27th Workshop \"What Comes Beyond the Standard Models\" Bled, July 8-17, 2024", "total_score": 0.5510714285714285, "query_match": 0.2714285714285714, "recency": 1.0, "study_type_priority": 0.5, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2501.11667", "title": "A One-Dimensional Energy Balance Model Parameterization for the Formation of CO2 Ice on the Surfaces", "total_score": 0.5510714285714285, "query_match": 0.2714285714285714, "recency": 1.0, "study_type_priority": 0.5, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 } ], "answer_summary": "๊ฒ€์ƒ‰ ์งˆ๋ฌธ: What are the technical details of RNN?\n\n๊ฒ€์ƒ‰ ๊ฒฐ๊ณผ ์š”์•ฝ:\n- ์ด 10๊ฐœ์˜ ๋…ผ๋ฌธ์ด ๊ฒ€์ƒ‰๋˜์—ˆ์Šต๋‹ˆ๋‹ค.\n- ์†Œ์Šค๋ณ„ ๋ถ„ํฌ: pubmed(7), arxiv(3)\n- ์—ฐ๋„ ๋ฒ”์œ„: 2021-2025\n- ํ‰๊ท  ๋žญํ‚น ์ ์ˆ˜: 0.578\n\n์ฃผ์š” ๋ฐœ๊ฒฌ:\n- ์ƒ์œ„ ๋…ผ๋ฌธ๋“ค์€ ์ฃผ๋กœ J Biomed Inform, Sci Rep, IEEE Trans Neural Netw Learn Syst์—์„œ ๋ฐœํ‘œ๋˜์—ˆ์Šต๋‹ˆ๋‹ค.\n- ์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ: systematic_review(1), experimental(5), theoretical(3), survey(1)", "charts": { "year_trend": "์—ฐ๋„๋ณ„ ํŠธ๋ Œ๋“œ ์ฐจํŠธ", "venue_distribution": "์ €๋„ ๋ถ„ํฌ ์ฐจํŠธ", "study_type_distribution": "์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ ์ฐจํŠธ" } }
What are the technical details of RNN?
10
๋‹ค์Œ ์งˆ๋ฌธ์— ๋Œ€ํ•ด ๋…ผ๋ฌธ์„ ๊ฒ€์ƒ‰ํ•˜๊ณ  ๋žญํ‚นํ•˜์—ฌ ๊ฒฐ๊ณผ๋ฅผ ์ œ์‹œํ•˜์„ธ์š”. ์งˆ๋ฌธ: What are the recent trends in biomedical? ์ถœ๋ ฅ ํ˜•์‹: 1. Evidence Table (Top-K): ๊ฒ€์ƒ‰๋œ ๋…ผ๋ฌธ์˜ ์ƒ์œ„ K๊ฐœ๋ฅผ ํ…Œ์ด๋ธ” ํ˜•์‹์œผ๋กœ ์ œ์‹œ 2. Ranking Breakdown: ๊ฐ ๋…ผ๋ฌธ์˜ ๋žญํ‚น ์ ์ˆ˜ ๊ตฌ์„ฑ ์š”์†Œ๋ฅผ ๊ณต๊ฐœ 3. Answer Summary: ๊ฒ€์ƒ‰ ๊ฒฐ๊ณผ๋ฅผ ์š”์•ฝํ•˜์—ฌ ์ œ์‹œ 4. Charts: ์—ฐ๋„๋ณ„ ํŠธ๋ Œ๋“œ, ์ €๋„ ๋ถ„ํฌ, ์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ ์ฐจํŠธ์— ๋Œ€ํ•œ ์„ค๋ช… ๋ชจ๋“  ์ฃผ์žฅ์€ Evidence Table์˜ ๋…ผ๋ฌธ์— ๊ทผ๊ฑฐํ•ด์•ผ ํ•˜๋ฉฐ, ๋žญํ‚น ๊ทผ๊ฑฐ๋ฅผ ๋ช…์‹œํ•ด์•ผ ํ•ฉ๋‹ˆ๋‹ค.
{ "evidence_table": [ { "rank": 1, "title": "Artificial Intelligence Tools in Biomedical Research: Part 1-Literature Search and Knowledge Mining.", "authors": "Sen Chandan K", "year": 2025, "venue": "Antioxid Redox Signal", "study_type": "systematic_review", "score": 0.6578571428571429, "url": "https://pubmed.ncbi.nlm.nih.gov/41433042/" }, { "rank": 2, "title": "What's Trending in Difference-in-Differences? A Synthesis of the Recent Econometrics Literature", "authors": "Jonathan Roth, Pedro H. C. Sant'Anna, Alyssa Bilinski", "year": 2022, "venue": "arXiv", "study_type": "experimental", "score": 0.6224999999999999, "url": "https://arxiv.org/abs/2201.01194v3" }, { "rank": 3, "title": "A Survey on Current Trends and Recent Advances in Text Anonymization", "authors": "Tobias DeuรŸer, Lorenz Sparrenberg, Armin Berger", "year": 2025, "venue": "arXiv", "study_type": "survey", "score": 0.6167857142857143, "url": "https://arxiv.org/abs/2508.21587v1" }, { "rank": 4, "title": "The National Poison data System<sup>ยฎ</sup>: a bibliographic review of publication trends and topic modeling using language models.", "authors": "Weigel Bernard, McCrary Quincy, Zweerink Kara", "year": 2025, "venue": "Clin Toxicol (Phila)", "study_type": "theoretical", "score": 0.6099999999999999, "url": "https://pubmed.ncbi.nlm.nih.gov/41432768/" }, { "rank": 5, "title": "Challenges of Trustworthy Federated Learning: What's Done, Current Trends and Remaining Work", "authors": "Nuria Rodrรญguez-Barroso, Mario Garcรญa-Mรกrquez, M. Victoria Luzรณn", "year": 2025, "venue": "arXiv", "study_type": "theoretical", "score": 0.6024999999999999, "url": "https://arxiv.org/abs/2507.15796v1" }, { "rank": 6, "title": "Systematic Review: Proteomics-Driven Multi-Omics Integration for Alzheimer's Disease Pathology and Precision Medicine.", "authors": "Dong Jonathan Mingsong, Zhong Huan", "year": 2025, "venue": "Neurol Int", "study_type": "systematic_review", "score": 0.6007142857142858, "url": "https://pubmed.ncbi.nlm.nih.gov/41441216/" }, { "rank": 7, "title": "Five decades of total ankle replacement: from early failures to fourth-generation innovations and future priorities.", "authors": "Colรฒ Gabriele, Fusini Federico, Mazzotti Antonio", "year": 2025, "venue": "Eur J Orthop Surg Traumatol", "study_type": "systematic_review", "score": 0.5985714285714285, "url": "https://pubmed.ncbi.nlm.nih.gov/41460300/" }, { "rank": 8, "title": "What drives Malaysian medical students' specialty choices? Findings from the MMSCI study.", "authors": "Ser Gerald Tze Zhen, Teng Zhi Sean, Hong Wei-Han", "year": 2025, "venue": "Hum Resour Health", "study_type": "survey", "score": 0.5978571428571429, "url": "https://pubmed.ncbi.nlm.nih.gov/41454333/" }, { "rank": 9, "title": "The research trends of macrophages in acute kidney injury: A bibliometric analysis.", "authors": "Zhang Xiangyu, Zhou Qiao, Bao Sizeng", "year": 2025, "venue": "Medicine (Baltimore)", "study_type": "other", "score": 0.5957142857142856, "url": "https://pubmed.ncbi.nlm.nih.gov/41465886/" }, { "rank": 10, "title": "Past, Present, and Future of Sodium Hypochlorite in Dermatology: A Scoping Review.", "authors": "Chang Christy H, Devine Maura, Woo Deborah", "year": 2025, "venue": "Am J Clin Dermatol", "study_type": "systematic_review", "score": 0.5950000000000001, "url": "https://pubmed.ncbi.nlm.nih.gov/41428161/" } ], "ranking_breakdown": [ { "paper_id": "41433042", "title": "Artificial Intelligence Tools in Biomedical Research: Part 1-Literature Search and Knowledge Mining.", "total_score": 0.6578571428571429, "query_match": 0.3571428571428571, "recency": 1.0, "study_type_priority": 0.9, "evidence_completeness": 0.8999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "2201.01194", "title": "What's Trending in Difference-in-Differences? A Synthesis of the Recent Econometrics Literature", "total_score": 0.6224999999999999, "query_match": 0.5, "recency": 0.7, "study_type_priority": 0.7, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2508.21587", "title": "A Survey on Current Trends and Recent Advances in Text Anonymization", "total_score": 0.6167857142857143, "query_match": 0.3857142857142857, "recency": 1.0, "study_type_priority": 0.6, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "41432768", "title": "The National Poison data System<sup>ยฎ</sup>: a bibliographic review of publication trends and topic ", "total_score": 0.6099999999999999, "query_match": 0.39999999999999997, "recency": 1.0, "study_type_priority": 0.5, "evidence_completeness": 0.9999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "2507.15796", "title": "Challenges of Trustworthy Federated Learning: What's Done, Current Trends and Remaining Work", "total_score": 0.6024999999999999, "query_match": 0.39999999999999997, "recency": 1.0, "study_type_priority": 0.5, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "41441216", "title": "Systematic Review: Proteomics-Driven Multi-Omics Integration for Alzheimer's Disease Pathology and P", "total_score": 0.6007142857142858, "query_match": 0.21428571428571427, "recency": 1.0, "study_type_priority": 0.9, "evidence_completeness": 0.8999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "41460300", "title": "Five decades of total ankle replacement: from early failures to fourth-generation innovations and fu", "total_score": 0.5985714285714285, "query_match": 0.1714285714285714, "recency": 1.0, "study_type_priority": 0.9, "evidence_completeness": 0.9999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "41454333", "title": "What drives Malaysian medical students' specialty choices? Findings from the MMSCI study.", "total_score": 0.5978571428571429, "query_match": 0.3571428571428571, "recency": 1.0, "study_type_priority": 0.6, "evidence_completeness": 0.8999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "41465886", "title": "The research trends of macrophages in acute kidney injury: A bibliometric analysis.", "total_score": 0.5957142857142856, "query_match": 0.41428571428571426, "recency": 1.0, "study_type_priority": 0.4, "evidence_completeness": 0.9999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "41428161", "title": "Past, Present, and Future of Sodium Hypochlorite in Dermatology: A Scoping Review.", "total_score": 0.5950000000000001, "query_match": 0.19999999999999998, "recency": 1.0, "study_type_priority": 0.9, "evidence_completeness": 0.8999999999999999, "duplicate_penalty": 0.0 } ], "answer_summary": "๊ฒ€์ƒ‰ ์งˆ๋ฌธ: What are the recent trends in biomedical?\n\n๊ฒ€์ƒ‰ ๊ฒฐ๊ณผ ์š”์•ฝ:\n- ์ด 10๊ฐœ์˜ ๋…ผ๋ฌธ์ด ๊ฒ€์ƒ‰๋˜์—ˆ์Šต๋‹ˆ๋‹ค.\n- ์†Œ์Šค๋ณ„ ๋ถ„ํฌ: pubmed(7), arxiv(3)\n- ์—ฐ๋„ ๋ฒ”์œ„: 2022-2025\n- ํ‰๊ท  ๋žญํ‚น ์ ์ˆ˜: 0.610\n\n์ฃผ์š” ๋ฐœ๊ฒฌ:\n- ์ƒ์œ„ ๋…ผ๋ฌธ๋“ค์€ ์ฃผ๋กœ Antioxid Redox Signal, arXiv, arXiv์—์„œ ๋ฐœํ‘œ๋˜์—ˆ์Šต๋‹ˆ๋‹ค.\n- ์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ: systematic_review(4), experimental(1), survey(2), theoretical(2), other(1)", "charts": { "year_trend": "์—ฐ๋„๋ณ„ ํŠธ๋ Œ๋“œ ์ฐจํŠธ", "venue_distribution": "์ €๋„ ๋ถ„ํฌ ์ฐจํŠธ", "study_type_distribution": "์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ ์ฐจํŠธ" } }
What are the recent trends in biomedical?
10
๋‹ค์Œ ์งˆ๋ฌธ์— ๋Œ€ํ•ด ๋…ผ๋ฌธ์„ ๊ฒ€์ƒ‰ํ•˜๊ณ  ๋žญํ‚นํ•˜์—ฌ ๊ฒฐ๊ณผ๋ฅผ ์ œ์‹œํ•˜์„ธ์š”. ์งˆ๋ฌธ: How has transformer evolved over the past 5 years? ์ถœ๋ ฅ ํ˜•์‹: 1. Evidence Table (Top-K): ๊ฒ€์ƒ‰๋œ ๋…ผ๋ฌธ์˜ ์ƒ์œ„ K๊ฐœ๋ฅผ ํ…Œ์ด๋ธ” ํ˜•์‹์œผ๋กœ ์ œ์‹œ 2. Ranking Breakdown: ๊ฐ ๋…ผ๋ฌธ์˜ ๋žญํ‚น ์ ์ˆ˜ ๊ตฌ์„ฑ ์š”์†Œ๋ฅผ ๊ณต๊ฐœ 3. Answer Summary: ๊ฒ€์ƒ‰ ๊ฒฐ๊ณผ๋ฅผ ์š”์•ฝํ•˜์—ฌ ์ œ์‹œ 4. Charts: ์—ฐ๋„๋ณ„ ํŠธ๋ Œ๋“œ, ์ €๋„ ๋ถ„ํฌ, ์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ ์ฐจํŠธ์— ๋Œ€ํ•œ ์„ค๋ช… ๋ชจ๋“  ์ฃผ์žฅ์€ Evidence Table์˜ ๋…ผ๋ฌธ์— ๊ทผ๊ฑฐํ•ด์•ผ ํ•˜๋ฉฐ, ๋žญํ‚น ๊ทผ๊ฑฐ๋ฅผ ๋ช…์‹œํ•ด์•ผ ํ•ฉ๋‹ˆ๋‹ค.
{ "evidence_table": [ { "rank": 1, "title": "Decoding the Digital Pulse: Bibliometric Analysis of 25 Years in Digital Health Research Through the Journal of Medical Internet Research.", "authors": "Kaczmarczyk Robert, Wilhelm Theresa Isabelle, Roos Jonas", "year": 2024, "venue": "J Med Internet Res", "study_type": "theoretical", "score": 0.5544444444444445, "url": "https://pubmed.ncbi.nlm.nih.gov/39546778/" }, { "rank": 2, "title": "Full Stack Optimization of Transformer Inference: a Survey", "authors": "Sehoon Kim, Coleman Hooper, Thanakul Wattanawong", "year": 2023, "venue": "arXiv", "study_type": "case_study", "score": 0.5380555555555555, "url": "https://arxiv.org/abs/2302.14017v1" }, { "rank": 3, "title": "Evolutionary Trends in Decision Sciences Education Research from Simulation and Games to Big Data Analytics and Generative Artificial Intelligence.", "authors": "Akpan Ikpe Justice, Razavi Rouzbeh, Akpan Asuama A", "year": 2025, "venue": "Big Data", "study_type": "theoretical", "score": 0.5166666666666666, "url": "https://pubmed.ncbi.nlm.nih.gov/40019821/" }, { "rank": 4, "title": "MLP Can Be A Good Transformer Learner", "authors": "Sihao Lin, Pumeng Lyu, Dongrui Liu", "year": 2024, "venue": "arXiv", "study_type": "experimental", "score": 0.5113888888888889, "url": "https://arxiv.org/abs/2404.05657v1" }, { "rank": 5, "title": "SupEuclid: Extremely Simple, High Quality OoD Detection with Supervised Contrastive Learning and Euclidean Distance", "authors": "Jarrod Haas", "year": 2023, "venue": "arXiv", "study_type": "experimental", "score": 0.49583333333333335, "url": "https://arxiv.org/abs/2308.10973v1" }, { "rank": 6, "title": "The Evolved Transformer", "authors": "David R. So, Chen Liang, Quoc V. Le", "year": 2019, "venue": "arXiv", "study_type": "experimental", "score": 0.49583333333333335, "url": "https://arxiv.org/abs/1901.11117v4" }, { "rank": 7, "title": "The gauge theory dual of the bilayer XY model with second order Josephson coupling", "authors": "Pye Ton How, Sungkit Yip", "year": 2025, "venue": "arXiv", "study_type": "theoretical", "score": 0.4913888888888889, "url": "https://arxiv.org/abs/2507.19401v2" }, { "rank": 8, "title": "PIGEON: Predicting Image Geolocations", "authors": "Lukas Haas, Michal Skreta, Silas Alberti", "year": 2023, "venue": "arXiv", "study_type": "experimental", "score": 0.48250000000000004, "url": "https://arxiv.org/abs/2307.05845v6" }, { "rank": 9, "title": "Glance-and-Gaze Vision Transformer", "authors": "Qihang Yu, Yingda Xia, Yutong Bai", "year": 2021, "venue": "arXiv", "study_type": "experimental", "score": 0.47805555555555546, "url": "https://arxiv.org/abs/2106.02277v1" }, { "rank": 10, "title": "Big Data Science Over the Past Web", "authors": "Miguel Costa, Julien Masanรจs", "year": 2021, "venue": "arXiv", "study_type": "theoretical", "score": 0.4691666666666666, "url": "https://arxiv.org/abs/2108.01605v1" } ], "ranking_breakdown": [ { "paper_id": "39546778", "title": "Decoding the Digital Pulse: Bibliometric Analysis of 25 Years in Digital Health Research Through the", "total_score": 0.5544444444444445, "query_match": 0.3111111111111111, "recency": 0.9, "study_type_priority": 0.5, "evidence_completeness": 0.9999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "2302.14017", "title": "Full Stack Optimization of Transformer Inference: a Survey", "total_score": 0.5380555555555555, "query_match": 0.28888888888888886, "recency": 0.8, "study_type_priority": 0.6, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "40019821", "title": "Evolutionary Trends in Decision Sciences Education Research from Simulation and Games to Big Data An", "total_score": 0.5166666666666666, "query_match": 0.16666666666666666, "recency": 1.0, "study_type_priority": 0.5, "evidence_completeness": 0.9999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "2404.05657", "title": "MLP Can Be A Good Transformer Learner", "total_score": 0.5113888888888889, "query_match": 0.12222222222222222, "recency": 0.9, "study_type_priority": 0.7, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2308.10973", "title": "SupEuclid: Extremely Simple, High Quality OoD Detection with Supervised Contrastive Learning and Euc", "total_score": 0.49583333333333335, "query_match": 0.13333333333333333, "recency": 0.8, "study_type_priority": 0.7, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "1901.11117", "title": "The Evolved Transformer", "total_score": 0.49583333333333335, "query_match": 0.3333333333333333, "recency": 0.4, "study_type_priority": 0.7, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2507.19401", "title": "The gauge theory dual of the bilayer XY model with second order Josephson coupling", "total_score": 0.4913888888888889, "query_match": 0.12222222222222222, "recency": 1.0, "study_type_priority": 0.5, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2307.05845", "title": "PIGEON: Predicting Image Geolocations", "total_score": 0.48250000000000004, "query_match": 0.09999999999999999, "recency": 0.8, "study_type_priority": 0.7, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2106.02277", "title": "Glance-and-Gaze Vision Transformer", "total_score": 0.47805555555555546, "query_match": 0.18888888888888888, "recency": 0.6, "study_type_priority": 0.7, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2108.01605", "title": "Big Data Science Over the Past Web", "total_score": 0.4691666666666666, "query_match": 0.26666666666666666, "recency": 0.6, "study_type_priority": 0.5, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 } ], "answer_summary": "๊ฒ€์ƒ‰ ์งˆ๋ฌธ: How has transformer evolved over the past 5 years?\n\n๊ฒ€์ƒ‰ ๊ฒฐ๊ณผ ์š”์•ฝ:\n- ์ด 10๊ฐœ์˜ ๋…ผ๋ฌธ์ด ๊ฒ€์ƒ‰๋˜์—ˆ์Šต๋‹ˆ๋‹ค.\n- ์†Œ์Šค๋ณ„ ๋ถ„ํฌ: pubmed(2), arxiv(8)\n- ์—ฐ๋„ ๋ฒ”์œ„: 2019-2025\n- ํ‰๊ท  ๋žญํ‚น ์ ์ˆ˜: 0.503\n\n์ฃผ์š” ๋ฐœ๊ฒฌ:\n- ์ƒ์œ„ ๋…ผ๋ฌธ๋“ค์€ ์ฃผ๋กœ J Med Internet Res, arXiv, Big Data์—์„œ ๋ฐœํ‘œ๋˜์—ˆ์Šต๋‹ˆ๋‹ค.\n- ์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ: theoretical(4), case_study(1), experimental(5)", "charts": { "year_trend": "์—ฐ๋„๋ณ„ ํŠธ๋ Œ๋“œ ์ฐจํŠธ", "venue_distribution": "์ €๋„ ๋ถ„ํฌ ์ฐจํŠธ", "study_type_distribution": "์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ ์ฐจํŠธ" } }
How has transformer evolved over the past 5 years?
10
๋‹ค์Œ ์งˆ๋ฌธ์— ๋Œ€ํ•ด ๋…ผ๋ฌธ์„ ๊ฒ€์ƒ‰ํ•˜๊ณ  ๋žญํ‚นํ•˜์—ฌ ๊ฒฐ๊ณผ๋ฅผ ์ œ์‹œํ•˜์„ธ์š”. ์งˆ๋ฌธ: What are the advantages of attention over traditional approaches? ์ถœ๋ ฅ ํ˜•์‹: 1. Evidence Table (Top-K): ๊ฒ€์ƒ‰๋œ ๋…ผ๋ฌธ์˜ ์ƒ์œ„ K๊ฐœ๋ฅผ ํ…Œ์ด๋ธ” ํ˜•์‹์œผ๋กœ ์ œ์‹œ 2. Ranking Breakdown: ๊ฐ ๋…ผ๋ฌธ์˜ ๋žญํ‚น ์ ์ˆ˜ ๊ตฌ์„ฑ ์š”์†Œ๋ฅผ ๊ณต๊ฐœ 3. Answer Summary: ๊ฒ€์ƒ‰ ๊ฒฐ๊ณผ๋ฅผ ์š”์•ฝํ•˜์—ฌ ์ œ์‹œ 4. Charts: ์—ฐ๋„๋ณ„ ํŠธ๋ Œ๋“œ, ์ €๋„ ๋ถ„ํฌ, ์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ ์ฐจํŠธ์— ๋Œ€ํ•œ ์„ค๋ช… ๋ชจ๋“  ์ฃผ์žฅ์€ Evidence Table์˜ ๋…ผ๋ฌธ์— ๊ทผ๊ฑฐํ•ด์•ผ ํ•˜๋ฉฐ, ๋žญํ‚น ๊ทผ๊ฑฐ๋ฅผ ๋ช…์‹œํ•ด์•ผ ํ•ฉ๋‹ˆ๋‹ค.
{ "evidence_table": [ { "rank": 1, "title": "Rethinking Hip Surgery: A Systematic Review of Sparing Piriformis and Internus, Repairing Externus (SPAIRE) vs. Traditional Hemiarthroplasty Approaches.", "authors": "Raufi Muhammad Y, Shrestha Sagaurav, Hamsho Ward", "year": 2025, "venue": "Cureus", "study_type": "systematic_review", "score": 0.6883333333333334, "url": "https://pubmed.ncbi.nlm.nih.gov/40896070/" }, { "rank": 2, "title": "CCFormer: Cross-Modal Cross-Attention Transformer for Classification of Hyperspectral and LiDAR Data.", "authors": "Guo Hufeng, Tian Baohui, Liu Wenyi", "year": 2025, "venue": "Sensors (Basel)", "study_type": "experimental", "score": 0.6127777777777778, "url": "https://pubmed.ncbi.nlm.nih.gov/41012936/" }, { "rank": 3, "title": "Diverse behavior clustering of students on campus with macroscopic attention.", "authors": "Chen Wanghu, Wu Zongjuan, Zeng Siqi", "year": 2025, "venue": "Sci Rep", "study_type": "case_study", "score": 0.5927777777777777, "url": "https://pubmed.ncbi.nlm.nih.gov/40813806/" }, { "rank": 4, "title": "Pedestrian detection in aerial image based on convolutional neural network with attention mechanism and multi-scale prediction.", "authors": "Yang Jiaxi, Shen Jiaquan, Wang Shitong", "year": 2025, "venue": "Sci Rep", "study_type": "experimental", "score": 0.5772222222222222, "url": "https://pubmed.ncbi.nlm.nih.gov/41387459/" }, { "rank": 5, "title": "The advantages of TCM in the treatment of gynecologic malignancies.", "authors": "Jiang Wei, Yuan Jinghua, Zhang Xinling", "year": 2025, "venue": "Front Oncol", "study_type": "other", "score": 0.5750000000000001, "url": "https://pubmed.ncbi.nlm.nih.gov/40978061/" }, { "rank": 6, "title": "Advancing oral squamous cell carcinoma research: the evolving role of patient-derived cell lines, xenografts, and organoid models.", "authors": "Upadhyay Vinal, Das Sourav, Nehru Amisha", "year": 2025, "venue": "3 Biotech", "study_type": "theoretical", "score": 0.5727777777777778, "url": "https://pubmed.ncbi.nlm.nih.gov/40822425/" }, { "rank": 7, "title": "A modular deep learning pipeline for enhanced plane-wave beamforming and B-mode image quality.", "authors": "Hadri Hamza, Fail Abderahhim, Sadik Mohamed", "year": 2025, "venue": "Med Phys", "study_type": "experimental", "score": 0.5700000000000001, "url": "https://pubmed.ncbi.nlm.nih.gov/40698752/" }, { "rank": 8, "title": "Attention as a Hypernetwork", "authors": "Simon Schug, Seijin Kobayashi, Yassir Akram", "year": 2024, "venue": "arXiv", "study_type": "experimental", "score": 0.5647222222222222, "url": "https://arxiv.org/abs/2406.05816v4" }, { "rank": 9, "title": "Large-Scale Production and Therapeutic Evaluation of Exosomes for Cancer Treatment.", "authors": "Kimiz-Gebologlu Ilgin, Oncel Suphi S", "year": 2025, "venue": "Thorac Res Pract", "study_type": "theoretical", "score": 0.5638888888888889, "url": "https://pubmed.ncbi.nlm.nih.gov/41340272/" }, { "rank": 10, "title": "Adaptive demand forecasting framework with weighted ensemble of regression and machine learning models along life cycle variability.", "authors": "Hammam Islam M, El-Kharbotly Amin K, Sadek Yomna M", "year": 2025, "venue": "Sci Rep", "study_type": "theoretical", "score": 0.5638888888888889, "url": "https://pubmed.ncbi.nlm.nih.gov/41188434/" } ], "ranking_breakdown": [ { "paper_id": "40896070", "title": "Rethinking Hip Surgery: A Systematic Review of Sparing Piriformis and Internus, Repairing Externus (", "total_score": 0.6883333333333334, "query_match": 0.43333333333333335, "recency": 1.0, "study_type_priority": 0.9, "evidence_completeness": 0.8999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "41012936", "title": "CCFormer: Cross-Modal Cross-Attention Transformer for Classification of Hyperspectral and LiDAR Data", "total_score": 0.6127777777777778, "query_match": 0.34444444444444444, "recency": 1.0, "study_type_priority": 0.7, "evidence_completeness": 0.8999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "40813806", "title": "Diverse behavior clustering of students on campus with macroscopic attention.", "total_score": 0.5927777777777777, "query_match": 0.34444444444444444, "recency": 1.0, "study_type_priority": 0.6, "evidence_completeness": 0.8999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "41387459", "title": "Pedestrian detection in aerial image based on convolutional neural network with attention mechanism ", "total_score": 0.5772222222222222, "query_match": 0.25555555555555554, "recency": 1.0, "study_type_priority": 0.7, "evidence_completeness": 0.8999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "40978061", "title": "The advantages of TCM in the treatment of gynecologic malignancies.", "total_score": 0.5750000000000001, "query_match": 0.4, "recency": 1.0, "study_type_priority": 0.4, "evidence_completeness": 0.8999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "40822425", "title": "Advancing oral squamous cell carcinoma research: the evolving role of patient-derived cell lines, xe", "total_score": 0.5727777777777778, "query_match": 0.34444444444444444, "recency": 1.0, "study_type_priority": 0.5, "evidence_completeness": 0.8999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "40698752", "title": "A modular deep learning pipeline for enhanced plane-wave beamforming and B-mode image quality.", "total_score": 0.5700000000000001, "query_match": 0.19999999999999998, "recency": 1.0, "study_type_priority": 0.7, "evidence_completeness": 0.9999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "2406.05816", "title": "Attention as a Hypernetwork", "total_score": 0.5647222222222222, "query_match": 0.25555555555555554, "recency": 0.9, "study_type_priority": 0.7, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "41340272", "title": "Large-Scale Production and Therapeutic Evaluation of Exosomes for Cancer Treatment.", "total_score": 0.5638888888888889, "query_match": 0.3222222222222222, "recency": 1.0, "study_type_priority": 0.5, "evidence_completeness": 0.8999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "41188434", "title": "Adaptive demand forecasting framework with weighted ensemble of regression and machine learning mode", "total_score": 0.5638888888888889, "query_match": 0.3222222222222222, "recency": 1.0, "study_type_priority": 0.5, "evidence_completeness": 0.8999999999999999, "duplicate_penalty": 0.0 } ], "answer_summary": "๊ฒ€์ƒ‰ ์งˆ๋ฌธ: What are the advantages of attention over traditional approaches?\n\n๊ฒ€์ƒ‰ ๊ฒฐ๊ณผ ์š”์•ฝ:\n- ์ด 10๊ฐœ์˜ ๋…ผ๋ฌธ์ด ๊ฒ€์ƒ‰๋˜์—ˆ์Šต๋‹ˆ๋‹ค.\n- ์†Œ์Šค๋ณ„ ๋ถ„ํฌ: pubmed(9), arxiv(1)\n- ์—ฐ๋„ ๋ฒ”์œ„: 2024-2025\n- ํ‰๊ท  ๋žญํ‚น ์ ์ˆ˜: 0.588\n\n์ฃผ์š” ๋ฐœ๊ฒฌ:\n- ์ƒ์œ„ ๋…ผ๋ฌธ๋“ค์€ ์ฃผ๋กœ Cureus, Sensors (Basel), Sci Rep์—์„œ ๋ฐœํ‘œ๋˜์—ˆ์Šต๋‹ˆ๋‹ค.\n- ์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ: systematic_review(1), experimental(4), case_study(1), other(1), theoretical(3)", "charts": { "year_trend": "์—ฐ๋„๋ณ„ ํŠธ๋ Œ๋“œ ์ฐจํŠธ", "venue_distribution": "์ €๋„ ๋ถ„ํฌ ์ฐจํŠธ", "study_type_distribution": "์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ ์ฐจํŠธ" } }
What are the advantages of attention over traditional approaches?
10
๋‹ค์Œ ์งˆ๋ฌธ์— ๋Œ€ํ•ด ๋…ผ๋ฌธ์„ ๊ฒ€์ƒ‰ํ•˜๊ณ  ๋žญํ‚นํ•˜์—ฌ ๊ฒฐ๊ณผ๋ฅผ ์ œ์‹œํ•˜์„ธ์š”. ์งˆ๋ฌธ: Compare self-supervised vs semi-supervised for classification ์ถœ๋ ฅ ํ˜•์‹: 1. Evidence Table (Top-K): ๊ฒ€์ƒ‰๋œ ๋…ผ๋ฌธ์˜ ์ƒ์œ„ K๊ฐœ๋ฅผ ํ…Œ์ด๋ธ” ํ˜•์‹์œผ๋กœ ์ œ์‹œ 2. Ranking Breakdown: ๊ฐ ๋…ผ๋ฌธ์˜ ๋žญํ‚น ์ ์ˆ˜ ๊ตฌ์„ฑ ์š”์†Œ๋ฅผ ๊ณต๊ฐœ 3. Answer Summary: ๊ฒ€์ƒ‰ ๊ฒฐ๊ณผ๋ฅผ ์š”์•ฝํ•˜์—ฌ ์ œ์‹œ 4. Charts: ์—ฐ๋„๋ณ„ ํŠธ๋ Œ๋“œ, ์ €๋„ ๋ถ„ํฌ, ์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ ์ฐจํŠธ์— ๋Œ€ํ•œ ์„ค๋ช… ๋ชจ๋“  ์ฃผ์žฅ์€ Evidence Table์˜ ๋…ผ๋ฌธ์— ๊ทผ๊ฑฐํ•ด์•ผ ํ•˜๋ฉฐ, ๋žญํ‚น ๊ทผ๊ฑฐ๋ฅผ ๋ช…์‹œํ•ด์•ผ ํ•ฉ๋‹ˆ๋‹ค.
{ "evidence_table": [ { "rank": 1, "title": "Rethinking Semi-Supervised Node Classification with Self-Supervised Graph Clustering", "authors": "Songbo Wang, Renchi Yang, Yurui Lai", "year": 2025, "venue": "arXiv", "study_type": "experimental", "score": 0.6825, "url": "https://arxiv.org/abs/2511.19976v1" }, { "rank": 2, "title": "Systematic comparison of semi-supervised and self-supervised learning for medical image classification", "authors": "Zhe Huang, Ruijie Jiang, Shuchin Aeron", "year": 2023, "venue": "arXiv", "study_type": "experimental", "score": 0.6710714285714285, "url": "https://arxiv.org/abs/2307.08919v3" }, { "rank": 3, "title": "ModSSC: A Modular Framework for Semi-Supervised Classification on Heterogeneous Data", "authors": "Melvin Barbaux", "year": 2025, "venue": "arXiv", "study_type": "experimental", "score": 0.6653571428571428, "url": "https://arxiv.org/abs/2512.13228v1" }, { "rank": 4, "title": "Towards Label-efficient Automatic Diagnosis and Analysis: A Comprehensive Survey of Advanced Deep Learning-based Weakly-supervised, Semi-supervised and Self-supervised Techniques in Histopathological Image Analysis", "authors": "Linhao Qu, Siyu Liu, Xiaoyu Liu", "year": 2022, "venue": "arXiv", "study_type": "systematic_review", "score": 0.6167857142857143, "url": "https://arxiv.org/abs/2208.08789v2" }, { "rank": 5, "title": "Shifting to Machine Supervision: Annotation-Efficient Semi and Self-Supervised Learning for Automatic Medical Image Segmentation and Classification", "authors": "Pranav Singh, Raviteja Chukkapalli, Shravan Chaudhari", "year": 2023, "venue": "arXiv", "study_type": "other", "score": 0.6110714285714286, "url": "https://arxiv.org/abs/2311.10319v6" }, { "rank": 6, "title": "FreeMatch: Self-adaptive Thresholding for Semi-supervised Learning", "authors": "Yidong Wang, Hao Chen, Qiang Heng", "year": 2022, "venue": "arXiv", "study_type": "experimental", "score": 0.6053571428571428, "url": "https://arxiv.org/abs/2205.07246v3" }, { "rank": 7, "title": "Semi-supervised Cervical Segmentation on Ultrasound by A Dual Framework for Neural Networks", "authors": "Fangyijie Wang, Kathleen M. Curran, Guรฉnolรฉ Silvestre", "year": 2025, "venue": "arXiv", "study_type": "theoretical", "score": 0.5967857142857143, "url": "https://arxiv.org/abs/2503.17057v1" }, { "rank": 8, "title": "Self-supervised Mean Teacher for Semi-supervised Chest X-ray Classification", "authors": "Fengbei Liu, Yu Tian, Filipe R. Cordeiro", "year": 2021, "venue": "arXiv", "study_type": "theoretical", "score": 0.5910714285714285, "url": "https://arxiv.org/abs/2103.03629v3" }, { "rank": 9, "title": "InterLUDE: Interactions between Labeled and Unlabeled Data to Enhance Semi-Supervised Learning", "authors": "Zhe Huang, Xiaowei Yu, Dajiang Zhu", "year": 2024, "venue": "arXiv", "study_type": "experimental", "score": 0.5882142857142857, "url": "https://arxiv.org/abs/2403.10658v1" }, { "rank": 10, "title": "SelfMatch: Combining Contrastive Self-Supervision and Consistency for Semi-Supervised Learning", "authors": "Byoungjip Kim, Jinho Choo, Yeong-Dae Kwon", "year": 2021, "venue": "arXiv", "study_type": "experimental", "score": 0.5853571428571428, "url": "https://arxiv.org/abs/2101.06480v1" } ], "ranking_breakdown": [ { "paper_id": "2511.19976", "title": "Rethinking Semi-Supervised Node Classification with Self-Supervised Graph Clustering", "total_score": 0.6825, "query_match": 0.5, "recency": 1.0, "study_type_priority": 0.7, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2307.08919", "title": "Systematic comparison of semi-supervised and self-supervised learning for medical image classificati", "total_score": 0.6710714285714285, "query_match": 0.5714285714285714, "recency": 0.8, "study_type_priority": 0.7, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2512.13228", "title": "ModSSC: A Modular Framework for Semi-Supervised Classification on Heterogeneous Data", "total_score": 0.6653571428571428, "query_match": 0.4571428571428571, "recency": 1.0, "study_type_priority": 0.7, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2208.08789", "title": "Towards Label-efficient Automatic Diagnosis and Analysis: A Comprehensive Survey of Advanced Deep Le", "total_score": 0.6167857142857143, "query_match": 0.3857142857142857, "recency": 0.7, "study_type_priority": 0.9, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2311.10319", "title": "Shifting to Machine Supervision: Annotation-Efficient Semi and Self-Supervised Learning for Automati", "total_score": 0.6110714285714286, "query_match": 0.5714285714285714, "recency": 0.8, "study_type_priority": 0.4, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2205.07246", "title": "FreeMatch: Self-adaptive Thresholding for Semi-supervised Learning", "total_score": 0.6053571428571428, "query_match": 0.4571428571428571, "recency": 0.7, "study_type_priority": 0.7, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2503.17057", "title": "Semi-supervised Cervical Segmentation on Ultrasound by A Dual Framework for Neural Networks", "total_score": 0.5967857142857143, "query_match": 0.3857142857142857, "recency": 1.0, "study_type_priority": 0.5, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2103.03629", "title": "Self-supervised Mean Teacher for Semi-supervised Chest X-ray Classification", "total_score": 0.5910714285714285, "query_match": 0.5714285714285714, "recency": 0.6, "study_type_priority": 0.5, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2403.10658", "title": "InterLUDE: Interactions between Labeled and Unlabeled Data to Enhance Semi-Supervised Learning", "total_score": 0.5882142857142857, "query_match": 0.3142857142857143, "recency": 0.9, "study_type_priority": 0.7, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2101.06480", "title": "SelfMatch: Combining Contrastive Self-Supervision and Consistency for Semi-Supervised Learning", "total_score": 0.5853571428571428, "query_match": 0.4571428571428571, "recency": 0.6, "study_type_priority": 0.7, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 } ], "answer_summary": "๊ฒ€์ƒ‰ ์งˆ๋ฌธ: Compare self-supervised vs semi-supervised for classification\n\n๊ฒ€์ƒ‰ ๊ฒฐ๊ณผ ์š”์•ฝ:\n- ์ด 10๊ฐœ์˜ ๋…ผ๋ฌธ์ด ๊ฒ€์ƒ‰๋˜์—ˆ์Šต๋‹ˆ๋‹ค.\n- ์†Œ์Šค๋ณ„ ๋ถ„ํฌ: arxiv(10)\n- ์—ฐ๋„ ๋ฒ”์œ„: 2021-2025\n- ํ‰๊ท  ๋žญํ‚น ์ ์ˆ˜: 0.621\n\n์ฃผ์š” ๋ฐœ๊ฒฌ:\n- ์ƒ์œ„ ๋…ผ๋ฌธ๋“ค์€ ์ฃผ๋กœ arXiv, arXiv, arXiv์—์„œ ๋ฐœํ‘œ๋˜์—ˆ์Šต๋‹ˆ๋‹ค.\n- ์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ: experimental(6), systematic_review(1), other(1), theoretical(2)", "charts": { "year_trend": "์—ฐ๋„๋ณ„ ํŠธ๋ Œ๋“œ ์ฐจํŠธ", "venue_distribution": "์ €๋„ ๋ถ„ํฌ ์ฐจํŠธ", "study_type_distribution": "์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ ์ฐจํŠธ" } }
Compare self-supervised vs semi-supervised for classification
10
๋‹ค์Œ ์งˆ๋ฌธ์— ๋Œ€ํ•ด ๋…ผ๋ฌธ์„ ๊ฒ€์ƒ‰ํ•˜๊ณ  ๋žญํ‚นํ•˜์—ฌ ๊ฒฐ๊ณผ๋ฅผ ์ œ์‹œํ•˜์„ธ์š”. ์งˆ๋ฌธ: How does attention work in large language models? ์ถœ๋ ฅ ํ˜•์‹: 1. Evidence Table (Top-K): ๊ฒ€์ƒ‰๋œ ๋…ผ๋ฌธ์˜ ์ƒ์œ„ K๊ฐœ๋ฅผ ํ…Œ์ด๋ธ” ํ˜•์‹์œผ๋กœ ์ œ์‹œ 2. Ranking Breakdown: ๊ฐ ๋…ผ๋ฌธ์˜ ๋žญํ‚น ์ ์ˆ˜ ๊ตฌ์„ฑ ์š”์†Œ๋ฅผ ๊ณต๊ฐœ 3. Answer Summary: ๊ฒ€์ƒ‰ ๊ฒฐ๊ณผ๋ฅผ ์š”์•ฝํ•˜์—ฌ ์ œ์‹œ 4. Charts: ์—ฐ๋„๋ณ„ ํŠธ๋ Œ๋“œ, ์ €๋„ ๋ถ„ํฌ, ์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ ์ฐจํŠธ์— ๋Œ€ํ•œ ์„ค๋ช… ๋ชจ๋“  ์ฃผ์žฅ์€ Evidence Table์˜ ๋…ผ๋ฌธ์— ๊ทผ๊ฑฐํ•ด์•ผ ํ•˜๋ฉฐ, ๋žญํ‚น ๊ทผ๊ฑฐ๋ฅผ ๋ช…์‹œํ•ด์•ผ ํ•ฉ๋‹ˆ๋‹ค.
{ "evidence_table": [ { "rank": 1, "title": "Fine-tuning large language models in federated learning with fairness-aware prompt selection.", "authors": "Jiang Yalan, Li Zhongliang, Song Bin", "year": 2026, "venue": "Neural Netw", "study_type": "experimental", "score": 0.7150000000000001, "url": "https://pubmed.ncbi.nlm.nih.gov/41072284/" }, { "rank": 2, "title": "A systematic review of ethical considerations of large language models in healthcare and medicine.", "authors": "Fareed Muhammad, Fatima Madeeha, Uddin Jamal", "year": 2025, "venue": "Front Digit Health", "study_type": "systematic_review", "score": 0.7050000000000001, "url": "https://pubmed.ncbi.nlm.nih.gov/41019285/" }, { "rank": 3, "title": "The potential existential threat of large language models to online survey research.", "authors": "Westwood Sean J", "year": 2025, "venue": "Proc Natl Acad Sci U S A", "study_type": "survey", "score": 0.665, "url": "https://pubmed.ncbi.nlm.nih.gov/41264250/" }, { "rank": 4, "title": "Soft Inductive Bias Approach via Explicit Reasoning Perspectives in Inappropriate Utterance Detection Using Large Language Models", "authors": "Ju-Young Kim, Ji-Hong Park, Se-Yeon Lee", "year": 2025, "venue": "arXiv", "study_type": "experimental", "score": 0.6575, "url": "https://arxiv.org/abs/2512.08480v1" }, { "rank": 5, "title": "Preprocessing narrative texts in electronic medical records to identify hospital adverse events: A scoping review.", "authors": "Jafarpour Hamed, Wu Guosong, Cheligeer Cheligeer Ken", "year": 2025, "venue": "Artif Intell Med", "study_type": "systematic_review", "score": 0.655, "url": "https://pubmed.ncbi.nlm.nih.gov/41072367/" }, { "rank": 6, "title": "Large language models in radiology workflows: An exploratory study of generative AI for non-visual tasks in the German healthcare system.", "authors": "Steinhauser Stefanie, Welsch Sabrina", "year": 2025, "venue": "Health Policy", "study_type": "theoretical", "score": 0.655, "url": "https://pubmed.ncbi.nlm.nih.gov/40972429/" }, { "rank": 7, "title": "A Survey of AIOps in the Era of Large Language Models", "authors": "Lingzhe Zhang, Tong Jia, Mengxi Jia", "year": 2025, "venue": "arXiv", "study_type": "survey", "score": 0.6525, "url": "https://arxiv.org/abs/2507.12472v1" }, { "rank": 8, "title": "Feasibility evaluation of large language models in anesthesia-specific post-operative care instructions for total knee arthroplasty.", "authors": "Nagesh Dhruv, Keating Donald P, Divakaruni Raghu V", "year": 2026, "venue": "PEC Innov", "study_type": "theoretical", "score": 0.645, "url": "https://pubmed.ncbi.nlm.nih.gov/41403686/" }, { "rank": 9, "title": "Vaner2: towards more general biomedical named entity recognition using multi-task large language model encoders.", "authors": "Liu Yuxuan, Bian Junyi, Zhai Weiqi", "year": 2026, "venue": "Health Inf Sci Syst", "study_type": "experimental", "score": 0.635, "url": "https://pubmed.ncbi.nlm.nih.gov/41281609/" }, { "rank": 10, "title": "Unforgettable Generalization in Language Models", "authors": "Eric Zhang, Leshem Chosen, Jacob Andreas", "year": 2024, "venue": "arXiv", "study_type": "randomized_trial", "score": 0.6325000000000001, "url": "https://arxiv.org/abs/2409.02228v1" } ], "ranking_breakdown": [ { "paper_id": "41072284", "title": "Fine-tuning large language models in federated learning with fairness-aware prompt selection.", "total_score": 0.7150000000000001, "query_match": 0.5125, "recency": 1.1, "study_type_priority": 0.7, "evidence_completeness": 0.9999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "41019285", "title": "A systematic review of ethical considerations of large language models in healthcare and medicine.", "total_score": 0.7050000000000001, "query_match": 0.475, "recency": 1.0, "study_type_priority": 0.9, "evidence_completeness": 0.8999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "41264250", "title": "The potential existential threat of large language models to online survey research.", "total_score": 0.665, "query_match": 0.4875, "recency": 1.0, "study_type_priority": 0.6, "evidence_completeness": 0.9999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "2512.08480", "title": "Soft Inductive Bias Approach via Explicit Reasoning Perspectives in Inappropriate Utterance Detectio", "total_score": 0.6575, "query_match": 0.4375, "recency": 1.0, "study_type_priority": 0.7, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "41072367", "title": "Preprocessing narrative texts in electronic medical records to identify hospital adverse events: A s", "total_score": 0.655, "query_match": 0.3125, "recency": 1.0, "study_type_priority": 0.9, "evidence_completeness": 0.9999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "40972429", "title": "Large language models in radiology workflows: An exploratory study of generative AI for non-visual t", "total_score": 0.655, "query_match": 0.5125, "recency": 1.0, "study_type_priority": 0.5, "evidence_completeness": 0.9999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "2507.12472", "title": "A Survey of AIOps in the Era of Large Language Models", "total_score": 0.6525, "query_match": 0.475, "recency": 1.0, "study_type_priority": 0.6, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "41403686", "title": "Feasibility evaluation of large language models in anesthesia-specific post-operative care instructi", "total_score": 0.645, "query_match": 0.475, "recency": 1.1, "study_type_priority": 0.5, "evidence_completeness": 0.8999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "41281609", "title": "Vaner2: towards more general biomedical named entity recognition using multi-task large language mod", "total_score": 0.635, "query_match": 0.35, "recency": 1.1, "study_type_priority": 0.7, "evidence_completeness": 0.8999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "2409.02228", "title": "Unforgettable Generalization in Language Models", "total_score": 0.6325000000000001, "query_match": 0.375, "recency": 0.9, "study_type_priority": 0.8, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 } ], "answer_summary": "๊ฒ€์ƒ‰ ์งˆ๋ฌธ: How does attention work in large language models?\n\n๊ฒ€์ƒ‰ ๊ฒฐ๊ณผ ์š”์•ฝ:\n- ์ด 10๊ฐœ์˜ ๋…ผ๋ฌธ์ด ๊ฒ€์ƒ‰๋˜์—ˆ์Šต๋‹ˆ๋‹ค.\n- ์†Œ์Šค๋ณ„ ๋ถ„ํฌ: pubmed(7), arxiv(3)\n- ์—ฐ๋„ ๋ฒ”์œ„: 2024-2026\n- ํ‰๊ท  ๋žญํ‚น ์ ์ˆ˜: 0.662\n\n์ฃผ์š” ๋ฐœ๊ฒฌ:\n- ์ƒ์œ„ ๋…ผ๋ฌธ๋“ค์€ ์ฃผ๋กœ Neural Netw, Front Digit Health, Proc Natl Acad Sci U S A์—์„œ ๋ฐœํ‘œ๋˜์—ˆ์Šต๋‹ˆ๋‹ค.\n- ์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ: experimental(3), systematic_review(2), survey(2), theoretical(2), randomized_trial(1)", "charts": { "year_trend": "์—ฐ๋„๋ณ„ ํŠธ๋ Œ๋“œ ์ฐจํŠธ", "venue_distribution": "์ €๋„ ๋ถ„ํฌ ์ฐจํŠธ", "study_type_distribution": "์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ ์ฐจํŠธ" } }
How does attention work in large language models?
10
๋‹ค์Œ ์งˆ๋ฌธ์— ๋Œ€ํ•ด ๋…ผ๋ฌธ์„ ๊ฒ€์ƒ‰ํ•˜๊ณ  ๋žญํ‚นํ•˜์—ฌ ๊ฒฐ๊ณผ๋ฅผ ์ œ์‹œํ•˜์„ธ์š”. ์งˆ๋ฌธ: What do systematic reviews say about attention mechanism? ์ถœ๋ ฅ ํ˜•์‹: 1. Evidence Table (Top-K): ๊ฒ€์ƒ‰๋œ ๋…ผ๋ฌธ์˜ ์ƒ์œ„ K๊ฐœ๋ฅผ ํ…Œ์ด๋ธ” ํ˜•์‹์œผ๋กœ ์ œ์‹œ 2. Ranking Breakdown: ๊ฐ ๋…ผ๋ฌธ์˜ ๋žญํ‚น ์ ์ˆ˜ ๊ตฌ์„ฑ ์š”์†Œ๋ฅผ ๊ณต๊ฐœ 3. Answer Summary: ๊ฒ€์ƒ‰ ๊ฒฐ๊ณผ๋ฅผ ์š”์•ฝํ•˜์—ฌ ์ œ์‹œ 4. Charts: ์—ฐ๋„๋ณ„ ํŠธ๋ Œ๋“œ, ์ €๋„ ๋ถ„ํฌ, ์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ ์ฐจํŠธ์— ๋Œ€ํ•œ ์„ค๋ช… ๋ชจ๋“  ์ฃผ์žฅ์€ Evidence Table์˜ ๋…ผ๋ฌธ์— ๊ทผ๊ฑฐํ•ด์•ผ ํ•˜๋ฉฐ, ๋žญํ‚น ๊ทผ๊ฑฐ๋ฅผ ๋ช…์‹œํ•ด์•ผ ํ•ฉ๋‹ˆ๋‹ค.
{ "evidence_table": [ { "rank": 1, "title": "What Do We Mean When We Talk about Trust in Social Media? A Systematic Review", "authors": "Yixuan Zhang, Joseph D Gaggiano, Nutchanon Yongsatianchot", "year": 2023, "venue": "arXiv", "study_type": "systematic_review", "score": 0.6425000000000001, "url": "https://arxiv.org/abs/2302.03671v1" }, { "rank": 2, "title": "Trust and Trustworthiness from Human-Centered Perspective in HRI -- A Systematic Literature Review", "authors": "Debora Firmino de Souza, Sonia Sousa, Kadri Kristjuhan-Ling", "year": 2025, "venue": "arXiv", "study_type": "systematic_review", "score": 0.6075, "url": "https://arxiv.org/abs/2501.19323v1" }, { "rank": 3, "title": "Computational-Assisted Systematic Review and Meta-Analysis (CASMA): Effect of a Subclass of GnRH-a on Endometriosis Recurrence", "authors": "Sandro Tsang", "year": 2025, "venue": "arXiv", "study_type": "meta_analysis", "score": 0.5975, "url": "https://arxiv.org/abs/2509.16599v3" }, { "rank": 4, "title": "Exploring the use of a Large Language Model for data extraction in systematic reviews: a rapid feasibility study", "authors": "Lena Schmidt, Kaitlyn Hair, Sergio Graziosi", "year": 2024, "venue": "arXiv", "study_type": "systematic_review", "score": 0.5825, "url": "https://arxiv.org/abs/2405.14445v2" }, { "rank": 5, "title": "Machine Learning Applications to Diffuse Reflectance Spectroscopy in Optical Diagnosis; A Systematic Review", "authors": "Nicola Rossberg, Celina L. Li, Simone Innocente", "year": 2025, "venue": "arXiv", "study_type": "systematic_review", "score": 0.5625, "url": "https://arxiv.org/abs/2503.02905v1" }, { "rank": 6, "title": "Non-invasive Techniques for Flow Rate Measurement in Water Pipes: Protocol for a Systematic Review", "authors": "Juan Diego Belesaca, Fabian Astudillo Salinas", "year": 2025, "venue": "arXiv", "study_type": "systematic_review", "score": 0.5625, "url": "https://arxiv.org/abs/2512.09239v1" }, { "rank": 7, "title": "Attention-Based Applications in Extended Reality to Support Autistic Users: A Systematic Review", "authors": "Katherine Wang, Simon Julier, Youngjun Cho", "year": 2022, "venue": "arXiv", "study_type": "systematic_review", "score": 0.5425, "url": "https://arxiv.org/abs/2204.00719v1" }, { "rank": 8, "title": "Pay Attention to What You Need", "authors": "Yifei Gao, Shaohong Chen, Lei Wang", "year": 2023, "venue": "arXiv", "study_type": "experimental", "score": 0.5225, "url": "https://arxiv.org/abs/2307.13365v3" }, { "rank": 9, "title": "A Systematic Review of Computational Thinking in Early Ages", "authors": "Edelberto Franco Silva, Bruno Josรฉ Dembogurski, Gustavo Silva Semaan", "year": 2021, "venue": "arXiv", "study_type": "systematic_review", "score": 0.49749999999999994, "url": "https://arxiv.org/abs/2106.10275v1" }, { "rank": 10, "title": "What Aspects of Mobile Ads Do Users Care About? An Empirical Study of Mobile In-app Ad Reviews", "authors": "Jiaping Gui, Meiyappan Nagappan, William G. J. Halfond", "year": 2017, "venue": "arXiv", "study_type": "experimental", "score": 0.48249999999999993, "url": "https://arxiv.org/abs/1702.07681v1" } ], "ranking_breakdown": [ { "paper_id": "2302.03671", "title": "What Do We Mean When We Talk about Trust in Social Media? A Systematic Review", "total_score": 0.6425000000000001, "query_match": 0.4, "recency": 0.8, "study_type_priority": 0.9, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2501.19323", "title": "Trust and Trustworthiness from Human-Centered Perspective in HRI -- A Systematic Literature Review", "total_score": 0.6075, "query_match": 0.2125, "recency": 1.0, "study_type_priority": 0.9, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2509.16599", "title": "Computational-Assisted Systematic Review and Meta-Analysis (CASMA): Effect of a Subclass of GnRH-a o", "total_score": 0.5975, "query_match": 0.1375, "recency": 1.0, "study_type_priority": 1.0, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2405.14445", "title": "Exploring the use of a Large Language Model for data extraction in systematic reviews: a rapid feasi", "total_score": 0.5825, "query_match": 0.2, "recency": 0.9, "study_type_priority": 0.9, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2503.02905", "title": "Machine Learning Applications to Diffuse Reflectance Spectroscopy in Optical Diagnosis; A Systematic", "total_score": 0.5625, "query_match": 0.1, "recency": 1.0, "study_type_priority": 0.9, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2512.09239", "title": "Non-invasive Techniques for Flow Rate Measurement in Water Pipes: Protocol for a Systematic Review", "total_score": 0.5625, "query_match": 0.1, "recency": 1.0, "study_type_priority": 0.9, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2204.00719", "title": "Attention-Based Applications in Extended Reality to Support Autistic Users: A Systematic Review", "total_score": 0.5425, "query_match": 0.2, "recency": 0.7, "study_type_priority": 0.9, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2307.13365", "title": "Pay Attention to What You Need", "total_score": 0.5225, "query_match": 0.2, "recency": 0.8, "study_type_priority": 0.7, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2106.10275", "title": "A Systematic Review of Computational Thinking in Early Ages", "total_score": 0.49749999999999994, "query_match": 0.1375, "recency": 0.6, "study_type_priority": 0.9, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "1702.07681", "title": "What Aspects of Mobile Ads Do Users Care About? An Empirical Study of Mobile In-app Ad Reviews", "total_score": 0.48249999999999993, "query_match": 0.4, "recency": 0.19999999999999996, "study_type_priority": 0.7, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 } ], "answer_summary": "๊ฒ€์ƒ‰ ์งˆ๋ฌธ: What do systematic reviews say about attention mechanism?\n\n๊ฒ€์ƒ‰ ๊ฒฐ๊ณผ ์š”์•ฝ:\n- ์ด 10๊ฐœ์˜ ๋…ผ๋ฌธ์ด ๊ฒ€์ƒ‰๋˜์—ˆ์Šต๋‹ˆ๋‹ค.\n- ์†Œ์Šค๋ณ„ ๋ถ„ํฌ: arxiv(10)\n- ์—ฐ๋„ ๋ฒ”์œ„: 2017-2025\n- ํ‰๊ท  ๋žญํ‚น ์ ์ˆ˜: 0.560\n\n์ฃผ์š” ๋ฐœ๊ฒฌ:\n- ์ƒ์œ„ ๋…ผ๋ฌธ๋“ค์€ ์ฃผ๋กœ arXiv, arXiv, arXiv์—์„œ ๋ฐœํ‘œ๋˜์—ˆ์Šต๋‹ˆ๋‹ค.\n- ์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ: systematic_review(7), meta_analysis(1), experimental(2)", "charts": { "year_trend": "์—ฐ๋„๋ณ„ ํŠธ๋ Œ๋“œ ์ฐจํŠธ", "venue_distribution": "์ €๋„ ๋ถ„ํฌ ์ฐจํŠธ", "study_type_distribution": "์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ ์ฐจํŠธ" } }
What do systematic reviews say about attention mechanism?
10
๋‹ค์Œ ์งˆ๋ฌธ์— ๋Œ€ํ•ด ๋…ผ๋ฌธ์„ ๊ฒ€์ƒ‰ํ•˜๊ณ  ๋žญํ‚นํ•˜์—ฌ ๊ฒฐ๊ณผ๋ฅผ ์ œ์‹œํ•˜์„ธ์š”. ์งˆ๋ฌธ: Compare unsupervised vs supervised for classification ์ถœ๋ ฅ ํ˜•์‹: 1. Evidence Table (Top-K): ๊ฒ€์ƒ‰๋œ ๋…ผ๋ฌธ์˜ ์ƒ์œ„ K๊ฐœ๋ฅผ ํ…Œ์ด๋ธ” ํ˜•์‹์œผ๋กœ ์ œ์‹œ 2. Ranking Breakdown: ๊ฐ ๋…ผ๋ฌธ์˜ ๋žญํ‚น ์ ์ˆ˜ ๊ตฌ์„ฑ ์š”์†Œ๋ฅผ ๊ณต๊ฐœ 3. Answer Summary: ๊ฒ€์ƒ‰ ๊ฒฐ๊ณผ๋ฅผ ์š”์•ฝํ•˜์—ฌ ์ œ์‹œ 4. Charts: ์—ฐ๋„๋ณ„ ํŠธ๋ Œ๋“œ, ์ €๋„ ๋ถ„ํฌ, ์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ ์ฐจํŠธ์— ๋Œ€ํ•œ ์„ค๋ช… ๋ชจ๋“  ์ฃผ์žฅ์€ Evidence Table์˜ ๋…ผ๋ฌธ์— ๊ทผ๊ฑฐํ•ด์•ผ ํ•˜๋ฉฐ, ๋žญํ‚น ๊ทผ๊ฑฐ๋ฅผ ๋ช…์‹œํ•ด์•ผ ํ•ฉ๋‹ˆ๋‹ค.
{ "evidence_table": [ { "rank": 1, "title": "Unsupervised Neighborhood Propagation Kernel Layers for Semi-supervised Node Classification", "authors": "Sonny Achten, Francesco Tonin, Panagiotis Patrinos", "year": 2023, "venue": "arXiv", "study_type": "experimental", "score": 0.6558333333333334, "url": "https://arxiv.org/abs/2301.13764v3" }, { "rank": 2, "title": "Comparative Analysis of Unsupervised and Supervised Autoencoders for Nuclei Classification in Clear Cell Renal Cell Carcinoma Images", "authors": "Fatemeh Javadian, Zahra Aminparast, Johannes Stegmaier", "year": 2025, "venue": "arXiv", "study_type": "theoretical", "score": 0.6558333333333333, "url": "https://arxiv.org/abs/2504.03146v1" }, { "rank": 3, "title": "ModSSC: A Modular Framework for Semi-Supervised Classification on Heterogeneous Data", "authors": "Melvin Barbaux", "year": 2025, "venue": "arXiv", "study_type": "experimental", "score": 0.6425, "url": "https://arxiv.org/abs/2512.13228v1" }, { "rank": 4, "title": "Beyond Supervised vs. Unsupervised: Representative Benchmarking and Analysis of Image Representation Learning", "authors": "Matthew Gwilliam, Abhinav Shrivastava", "year": 2022, "venue": "arXiv", "study_type": "theoretical", "score": 0.6024999999999999, "url": "https://arxiv.org/abs/2206.08347v1" }, { "rank": 5, "title": "Bridge the Gap between Supervised and Unsupervised Learning for Fine-Grained Classification", "authors": "Jiabao Wang, Yang Li, Xiu-Shen Wei", "year": 2022, "venue": "arXiv", "study_type": "theoretical", "score": 0.5958333333333333, "url": "https://arxiv.org/abs/2203.00441v1" }, { "rank": 6, "title": "Unsupervised structure classes vs. supervised property classes of silicon quantum dots using neural networks.", "authors": "Parker Amanda J, Barnard Amanda S", "year": 2021, "venue": "Nanoscale Horiz", "study_type": "experimental", "score": 0.595, "url": "https://pubmed.ncbi.nlm.nih.gov/33527922/" }, { "rank": 7, "title": "Long-term foehn reconstruction combining unsupervised and supervised learning", "authors": "Reto Stauffer, Achim Zeileis, Georg J. Mayr", "year": 2024, "venue": "arXiv", "study_type": "theoretical", "score": 0.5691666666666666, "url": "https://arxiv.org/abs/2406.01818v2" }, { "rank": 8, "title": "Text classification in shipping industry using unsupervised models and Transformer based supervised models", "authors": "Ying Xie, Dongping Song", "year": 2022, "venue": "arXiv", "study_type": "theoretical", "score": 0.5624999999999999, "url": "https://arxiv.org/abs/2212.12407v1" }, { "rank": 9, "title": "Semi-supervised and Unsupervised Methods for Heart Sounds Classification in Restricted Data Environments", "authors": "Balagopal Unnikrishnan, Pranshu Ranjan Singh, Xulei Yang", "year": 2020, "venue": "arXiv", "study_type": "theoretical", "score": 0.5558333333333333, "url": "https://arxiv.org/abs/2006.02610v1" }, { "rank": 10, "title": "Shifting to Machine Supervision: Annotation-Efficient Semi and Self-Supervised Learning for Automatic Medical Image Segmentation and Classification", "authors": "Pranav Singh, Raviteja Chukkapalli, Shravan Chaudhari", "year": 2023, "venue": "arXiv", "study_type": "other", "score": 0.5425000000000001, "url": "https://arxiv.org/abs/2311.10319v6" } ], "ranking_breakdown": [ { "paper_id": "2301.13764", "title": "Unsupervised Neighborhood Propagation Kernel Layers for Semi-supervised Node Classification", "total_score": 0.6558333333333334, "query_match": 0.5333333333333333, "recency": 0.8, "study_type_priority": 0.7, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2504.03146", "title": "Comparative Analysis of Unsupervised and Supervised Autoencoders for Nuclei Classification in Clear ", "total_score": 0.6558333333333333, "query_match": 0.5333333333333333, "recency": 1.0, "study_type_priority": 0.5, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2512.13228", "title": "ModSSC: A Modular Framework for Semi-Supervised Classification on Heterogeneous Data", "total_score": 0.6425, "query_match": 0.4, "recency": 1.0, "study_type_priority": 0.7, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2206.08347", "title": "Beyond Supervised vs. Unsupervised: Representative Benchmarking and Analysis of Image Representation", "total_score": 0.6024999999999999, "query_match": 0.55, "recency": 0.7, "study_type_priority": 0.5, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2203.00441", "title": "Bridge the Gap between Supervised and Unsupervised Learning for Fine-Grained Classification", "total_score": 0.5958333333333333, "query_match": 0.5333333333333333, "recency": 0.7, "study_type_priority": 0.5, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "33527922", "title": "Unsupervised structure classes vs. supervised property classes of silicon quantum dots using neural ", "total_score": 0.595, "query_match": 0.5, "recency": 0.6, "study_type_priority": 0.7, "evidence_completeness": 0.8999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "2406.01818", "title": "Long-term foehn reconstruction combining unsupervised and supervised learning", "total_score": 0.5691666666666666, "query_match": 0.36666666666666664, "recency": 0.9, "study_type_priority": 0.5, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2212.12407", "title": "Text classification in shipping industry using unsupervised models and Transformer based supervised ", "total_score": 0.5624999999999999, "query_match": 0.44999999999999996, "recency": 0.7, "study_type_priority": 0.5, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2006.02610", "title": "Semi-supervised and Unsupervised Methods for Heart Sounds Classification in Restricted Data Environm", "total_score": 0.5558333333333333, "query_match": 0.5333333333333333, "recency": 0.5, "study_type_priority": 0.5, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2311.10319", "title": "Shifting to Machine Supervision: Annotation-Efficient Semi and Self-Supervised Learning for Automati", "total_score": 0.5425000000000001, "query_match": 0.4, "recency": 0.8, "study_type_priority": 0.4, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 } ], "answer_summary": "๊ฒ€์ƒ‰ ์งˆ๋ฌธ: Compare unsupervised vs supervised for classification\n\n๊ฒ€์ƒ‰ ๊ฒฐ๊ณผ ์š”์•ฝ:\n- ์ด 10๊ฐœ์˜ ๋…ผ๋ฌธ์ด ๊ฒ€์ƒ‰๋˜์—ˆ์Šต๋‹ˆ๋‹ค.\n- ์†Œ์Šค๋ณ„ ๋ถ„ํฌ: arxiv(9), pubmed(1)\n- ์—ฐ๋„ ๋ฒ”์œ„: 2020-2025\n- ํ‰๊ท  ๋žญํ‚น ์ ์ˆ˜: 0.598\n\n์ฃผ์š” ๋ฐœ๊ฒฌ:\n- ์ƒ์œ„ ๋…ผ๋ฌธ๋“ค์€ ์ฃผ๋กœ arXiv, arXiv, arXiv์—์„œ ๋ฐœํ‘œ๋˜์—ˆ์Šต๋‹ˆ๋‹ค.\n- ์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ: experimental(3), theoretical(6), other(1)", "charts": { "year_trend": "์—ฐ๋„๋ณ„ ํŠธ๋ Œ๋“œ ์ฐจํŠธ", "venue_distribution": "์ €๋„ ๋ถ„ํฌ ์ฐจํŠธ", "study_type_distribution": "์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ ์ฐจํŠธ" } }
Compare unsupervised vs supervised for classification
10
๋‹ค์Œ ์งˆ๋ฌธ์— ๋Œ€ํ•ด ๋…ผ๋ฌธ์„ ๊ฒ€์ƒ‰ํ•˜๊ณ  ๋žญํ‚นํ•˜์—ฌ ๊ฒฐ๊ณผ๋ฅผ ์ œ์‹œํ•˜์„ธ์š”. ์งˆ๋ฌธ: What do systematic reviews say about deep learning? ์ถœ๋ ฅ ํ˜•์‹: 1. Evidence Table (Top-K): ๊ฒ€์ƒ‰๋œ ๋…ผ๋ฌธ์˜ ์ƒ์œ„ K๊ฐœ๋ฅผ ํ…Œ์ด๋ธ” ํ˜•์‹์œผ๋กœ ์ œ์‹œ 2. Ranking Breakdown: ๊ฐ ๋…ผ๋ฌธ์˜ ๋žญํ‚น ์ ์ˆ˜ ๊ตฌ์„ฑ ์š”์†Œ๋ฅผ ๊ณต๊ฐœ 3. Answer Summary: ๊ฒ€์ƒ‰ ๊ฒฐ๊ณผ๋ฅผ ์š”์•ฝํ•˜์—ฌ ์ œ์‹œ 4. Charts: ์—ฐ๋„๋ณ„ ํŠธ๋ Œ๋“œ, ์ €๋„ ๋ถ„ํฌ, ์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ ์ฐจํŠธ์— ๋Œ€ํ•œ ์„ค๋ช… ๋ชจ๋“  ์ฃผ์žฅ์€ Evidence Table์˜ ๋…ผ๋ฌธ์— ๊ทผ๊ฑฐํ•ด์•ผ ํ•˜๋ฉฐ, ๋žญํ‚น ๊ทผ๊ฑฐ๋ฅผ ๋ช…์‹œํ•ด์•ผ ํ•ฉ๋‹ˆ๋‹ค.
{ "evidence_table": [ { "rank": 1, "title": "What Do We Mean When We Talk about Trust in Social Media? A Systematic Review", "authors": "Yixuan Zhang, Joseph D Gaggiano, Nutchanon Yongsatianchot", "year": 2023, "venue": "arXiv", "study_type": "systematic_review", "score": 0.6425000000000001, "url": "https://arxiv.org/abs/2302.03671v1" }, { "rank": 2, "title": "Innovative Speech-Based Deep Learning Approaches for Parkinson's Disease Classification: A Systematic Review", "authors": "Lisanne van Gelderen, Cristian Tejedor-Garcรญa", "year": 2024, "venue": "arXiv", "study_type": "systematic_review", "score": 0.6225, "url": "https://arxiv.org/abs/2407.17844v4" }, { "rank": 3, "title": "Deep Learning for Computer Vision based Activity Recognition and Fall Detection of the Elderly: a Systematic Review", "authors": "F. Xavier Gaya-Morey, Cristina Manresa-Yee, Jose M. Buades-Rubio", "year": 2024, "venue": "arXiv", "study_type": "systematic_review", "score": 0.6225, "url": "https://arxiv.org/abs/2401.11790v3" }, { "rank": 4, "title": "Trust and Trustworthiness from Human-Centered Perspective in HRI -- A Systematic Literature Review", "authors": "Debora Firmino de Souza, Sonia Sousa, Kadri Kristjuhan-Ling", "year": 2025, "venue": "arXiv", "study_type": "systematic_review", "score": 0.6075, "url": "https://arxiv.org/abs/2501.19323v1" }, { "rank": 5, "title": "Exploring the use of a Large Language Model for data extraction in systematic reviews: a rapid feasibility study", "authors": "Lena Schmidt, Kaitlyn Hair, Sergio Graziosi", "year": 2024, "venue": "arXiv", "study_type": "systematic_review", "score": 0.5825, "url": "https://arxiv.org/abs/2405.14445v2" }, { "rank": 6, "title": "Hybrid deep learning-based strategy for the hepatocellular carcinoma cancer grade classification of H&E stained liver histopathology images", "authors": "Ajinkya Deshpande, Deep Gupta, Ankit Bhurane", "year": 2024, "venue": "arXiv", "study_type": "theoretical", "score": 0.5025, "url": "https://arxiv.org/abs/2412.03084v2" }, { "rank": 7, "title": "A Systematic Review of Computational Thinking in Early Ages", "authors": "Edelberto Franco Silva, Bruno Josรฉ Dembogurski, Gustavo Silva Semaan", "year": 2021, "venue": "arXiv", "study_type": "systematic_review", "score": 0.49749999999999994, "url": "https://arxiv.org/abs/2106.10275v1" }, { "rank": 8, "title": "Deep Learning and Computational Physics (Lecture Notes)", "authors": "Deep Ray, Orazio Pinti, Assad A. Oberai", "year": 2023, "venue": "arXiv", "study_type": "theoretical", "score": 0.48250000000000004, "url": "https://arxiv.org/abs/2301.00942v1" }, { "rank": 9, "title": "Deep Generative Models for Detector Signature Simulation: A Taxonomic Review", "authors": "Baran Hashemi, Claudius Krause", "year": 2023, "venue": "arXiv", "study_type": "experimental", "score": 0.48250000000000004, "url": "https://arxiv.org/abs/2312.09597v2" }, { "rank": 10, "title": "Deep Video Codec Control for Vision Models", "authors": "Christoph Reich, Biplob Debnath, Deep Patel", "year": 2023, "venue": "arXiv", "study_type": "experimental", "score": 0.48250000000000004, "url": "https://arxiv.org/abs/2308.16215v6" } ], "ranking_breakdown": [ { "paper_id": "2302.03671", "title": "What Do We Mean When We Talk about Trust in Social Media? A Systematic Review", "total_score": 0.6425000000000001, "query_match": 0.4, "recency": 0.8, "study_type_priority": 0.9, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2407.17844", "title": "Innovative Speech-Based Deep Learning Approaches for Parkinson's Disease Classification: A Systemati", "total_score": 0.6225, "query_match": 0.3, "recency": 0.9, "study_type_priority": 0.9, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2401.11790", "title": "Deep Learning for Computer Vision based Activity Recognition and Fall Detection of the Elderly: a Sy", "total_score": 0.6225, "query_match": 0.3, "recency": 0.9, "study_type_priority": 0.9, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2501.19323", "title": "Trust and Trustworthiness from Human-Centered Perspective in HRI -- A Systematic Literature Review", "total_score": 0.6075, "query_match": 0.2125, "recency": 1.0, "study_type_priority": 0.9, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2405.14445", "title": "Exploring the use of a Large Language Model for data extraction in systematic reviews: a rapid feasi", "total_score": 0.5825, "query_match": 0.2, "recency": 0.9, "study_type_priority": 0.9, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2412.03084", "title": "Hybrid deep learning-based strategy for the hepatocellular carcinoma cancer grade classification of ", "total_score": 0.5025, "query_match": 0.2, "recency": 0.9, "study_type_priority": 0.5, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2106.10275", "title": "A Systematic Review of Computational Thinking in Early Ages", "total_score": 0.49749999999999994, "query_match": 0.1375, "recency": 0.6, "study_type_priority": 0.9, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2301.00942", "title": "Deep Learning and Computational Physics (Lecture Notes)", "total_score": 0.48250000000000004, "query_match": 0.2, "recency": 0.8, "study_type_priority": 0.5, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2312.09597", "title": "Deep Generative Models for Detector Signature Simulation: A Taxonomic Review", "total_score": 0.48250000000000004, "query_match": 0.1, "recency": 0.8, "study_type_priority": 0.7, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2308.16215", "title": "Deep Video Codec Control for Vision Models", "total_score": 0.48250000000000004, "query_match": 0.1, "recency": 0.8, "study_type_priority": 0.7, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 } ], "answer_summary": "๊ฒ€์ƒ‰ ์งˆ๋ฌธ: What do systematic reviews say about deep learning?\n\n๊ฒ€์ƒ‰ ๊ฒฐ๊ณผ ์š”์•ฝ:\n- ์ด 10๊ฐœ์˜ ๋…ผ๋ฌธ์ด ๊ฒ€์ƒ‰๋˜์—ˆ์Šต๋‹ˆ๋‹ค.\n- ์†Œ์Šค๋ณ„ ๋ถ„ํฌ: arxiv(10)\n- ์—ฐ๋„ ๋ฒ”์œ„: 2021-2025\n- ํ‰๊ท  ๋žญํ‚น ์ ์ˆ˜: 0.552\n\n์ฃผ์š” ๋ฐœ๊ฒฌ:\n- ์ƒ์œ„ ๋…ผ๋ฌธ๋“ค์€ ์ฃผ๋กœ arXiv, arXiv, arXiv์—์„œ ๋ฐœํ‘œ๋˜์—ˆ์Šต๋‹ˆ๋‹ค.\n- ์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ: systematic_review(6), theoretical(2), experimental(2)", "charts": { "year_trend": "์—ฐ๋„๋ณ„ ํŠธ๋ Œ๋“œ ์ฐจํŠธ", "venue_distribution": "์ €๋„ ๋ถ„ํฌ ์ฐจํŠธ", "study_type_distribution": "์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ ์ฐจํŠธ" } }
What do systematic reviews say about deep learning?
10
๋‹ค์Œ ์งˆ๋ฌธ์— ๋Œ€ํ•ด ๋…ผ๋ฌธ์„ ๊ฒ€์ƒ‰ํ•˜๊ณ  ๋žญํ‚นํ•˜์—ฌ ๊ฒฐ๊ณผ๋ฅผ ์ œ์‹œํ•˜์„ธ์š”. ์งˆ๋ฌธ: What are the emerging topics in large language models? ์ถœ๋ ฅ ํ˜•์‹: 1. Evidence Table (Top-K): ๊ฒ€์ƒ‰๋œ ๋…ผ๋ฌธ์˜ ์ƒ์œ„ K๊ฐœ๋ฅผ ํ…Œ์ด๋ธ” ํ˜•์‹์œผ๋กœ ์ œ์‹œ 2. Ranking Breakdown: ๊ฐ ๋…ผ๋ฌธ์˜ ๋žญํ‚น ์ ์ˆ˜ ๊ตฌ์„ฑ ์š”์†Œ๋ฅผ ๊ณต๊ฐœ 3. Answer Summary: ๊ฒ€์ƒ‰ ๊ฒฐ๊ณผ๋ฅผ ์š”์•ฝํ•˜์—ฌ ์ œ์‹œ 4. Charts: ์—ฐ๋„๋ณ„ ํŠธ๋ Œ๋“œ, ์ €๋„ ๋ถ„ํฌ, ์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ ์ฐจํŠธ์— ๋Œ€ํ•œ ์„ค๋ช… ๋ชจ๋“  ์ฃผ์žฅ์€ Evidence Table์˜ ๋…ผ๋ฌธ์— ๊ทผ๊ฑฐํ•ด์•ผ ํ•˜๋ฉฐ, ๋žญํ‚น ๊ทผ๊ฑฐ๋ฅผ ๋ช…์‹œํ•ด์•ผ ํ•ฉ๋‹ˆ๋‹ค.
{ "evidence_table": [ { "rank": 1, "title": "Comparison of large language models in oral and maxillofacial surgery.", "authors": "Grillo Ricardo, Llanos Alexandre Hugo, Costa Claudio", "year": 2025, "venue": "Br J Oral Maxillofac Surg", "study_type": "systematic_review", "score": 0.6972222222222223, "url": "https://pubmed.ncbi.nlm.nih.gov/41076417/" }, { "rank": 2, "title": "Small versus large bore chest tube in traumatic hemothorax, hemopneumothorax, and pneumothorax: a meta-analysis of randomized controlled trials with trial sequential analysis.", "authors": "Granieri Stefano, Cioffi Stefano Piero Bernardo, Asaro Alessandro", "year": 2025, "venue": "World J Emerg Surg", "study_type": "meta_analysis", "score": 0.6877777777777778, "url": "https://pubmed.ncbi.nlm.nih.gov/41287089/" }, { "rank": 3, "title": "Physician Perspectives on Large Language Models in Health Care: A Cross-Sectional Survey Study.", "authors": "Hong Hyo Jung, Shah Nigam H, Pfeffer Michael A", "year": 2025, "venue": "Appl Clin Inform", "study_type": "survey", "score": 0.6655555555555555, "url": "https://pubmed.ncbi.nlm.nih.gov/41167595/" }, { "rank": 4, "title": "Large language models in clinical trials: applications, technical advances, and future directions.", "authors": "Lin Anqi, Wang Zhihan, Jiang Aimin", "year": 2025, "venue": "BMC Med", "study_type": "theoretical", "score": 0.6588888888888889, "url": "https://pubmed.ncbi.nlm.nih.gov/41088200/" }, { "rank": 5, "title": "Integrating clinical guidelines with large language models for improved sepsis mortality prediction.", "authors": "Zhao Zhen, An Bo, Zhang Tianpeng", "year": 2025, "venue": "Health Informatics J", "study_type": "experimental", "score": 0.6499999999999999, "url": "https://pubmed.ncbi.nlm.nih.gov/41195684/" }, { "rank": 6, "title": "Large language models as educational collaborators: developing non-conventional teaching aids in pharmacology & therapeutics.", "authors": "Sridharan Kannan, Sivaramakrishnan Gowri", "year": 2025, "venue": "BMC Med Educ", "study_type": "theoretical", "score": 0.6455555555555554, "url": "https://pubmed.ncbi.nlm.nih.gov/41168789/" }, { "rank": 7, "title": "Soft Inductive Bias Approach via Explicit Reasoning Perspectives in Inappropriate Utterance Detection Using Large Language Models", "authors": "Ju-Young Kim, Ji-Hong Park, Se-Yeon Lee", "year": 2025, "venue": "arXiv", "study_type": "experimental", "score": 0.6380555555555555, "url": "https://arxiv.org/abs/2512.08480v1" }, { "rank": 8, "title": "Evaluating Large Language Models for Sentiment Analysis and Hesitancy Analysis on Vaccine Posts From Social Media: Qualitative Study.", "authors": "Annan Augustine, Eiden Amanda L, Wang Dong", "year": 2025, "venue": "JMIR Form Res", "study_type": "theoretical", "score": 0.6366666666666667, "url": "https://pubmed.ncbi.nlm.nih.gov/41092067/" }, { "rank": 9, "title": "Large language models for ESC guideline interpretation: a targeted review of accuracy and applicability.", "authors": "Olariu Maria-Ecaterina, Burlacu Alexandru, Brinza Crischentian", "year": 2025, "venue": "Future Cardiol", "study_type": "theoretical", "score": 0.6366666666666667, "url": "https://pubmed.ncbi.nlm.nih.gov/41074629/" }, { "rank": 10, "title": "Assessing Large Language Models in Building a Structured Dataset From AskDocs Subreddit Data: Methodological Study.", "authors": "Snell Quinn, Westhoff Chase, Westhoff John", "year": 2025, "venue": "J Med Internet Res", "study_type": "theoretical", "score": 0.6322222222222222, "url": "https://pubmed.ncbi.nlm.nih.gov/41124662/" } ], "ranking_breakdown": [ { "paper_id": "41076417", "title": "Comparison of large language models in oral and maxillofacial surgery.", "total_score": 0.6972222222222223, "query_match": 0.45555555555555555, "recency": 1.0, "study_type_priority": 0.9, "evidence_completeness": 0.8999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "41287089", "title": "Small versus large bore chest tube in traumatic hemothorax, hemopneumothorax, and pneumothorax: a me", "total_score": 0.6877777777777778, "query_match": 0.34444444444444444, "recency": 1.0, "study_type_priority": 1.0, "evidence_completeness": 0.9999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "41167595", "title": "Physician Perspectives on Large Language Models in Health Care: A Cross-Sectional Survey Study.", "total_score": 0.6655555555555555, "query_match": 0.4888888888888888, "recency": 1.0, "study_type_priority": 0.6, "evidence_completeness": 0.9999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "41088200", "title": "Large language models in clinical trials: applications, technical advances, and future directions.", "total_score": 0.6588888888888889, "query_match": 0.5222222222222223, "recency": 1.0, "study_type_priority": 0.5, "evidence_completeness": 0.9999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "41195684", "title": "Integrating clinical guidelines with large language models for improved sepsis mortality prediction.", "total_score": 0.6499999999999999, "query_match": 0.39999999999999997, "recency": 1.0, "study_type_priority": 0.7, "evidence_completeness": 0.9999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "41168789", "title": "Large language models as educational collaborators: developing non-conventional teaching aids in pha", "total_score": 0.6455555555555554, "query_match": 0.4888888888888888, "recency": 1.0, "study_type_priority": 0.5, "evidence_completeness": 0.9999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "2512.08480", "title": "Soft Inductive Bias Approach via Explicit Reasoning Perspectives in Inappropriate Utterance Detectio", "total_score": 0.6380555555555555, "query_match": 0.38888888888888884, "recency": 1.0, "study_type_priority": 0.7, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "41092067", "title": "Evaluating Large Language Models for Sentiment Analysis and Hesitancy Analysis on Vaccine Posts From", "total_score": 0.6366666666666667, "query_match": 0.4666666666666667, "recency": 1.0, "study_type_priority": 0.5, "evidence_completeness": 0.9999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "41074629", "title": "Large language models for ESC guideline interpretation: a targeted review of accuracy and applicabil", "total_score": 0.6366666666666667, "query_match": 0.4666666666666667, "recency": 1.0, "study_type_priority": 0.5, "evidence_completeness": 0.9999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "41124662", "title": "Assessing Large Language Models in Building a Structured Dataset From AskDocs Subreddit Data: Method", "total_score": 0.6322222222222222, "query_match": 0.4555555555555555, "recency": 1.0, "study_type_priority": 0.5, "evidence_completeness": 0.9999999999999999, "duplicate_penalty": 0.0 } ], "answer_summary": "๊ฒ€์ƒ‰ ์งˆ๋ฌธ: What are the emerging topics in large language models?\n\n๊ฒ€์ƒ‰ ๊ฒฐ๊ณผ ์š”์•ฝ:\n- ์ด 10๊ฐœ์˜ ๋…ผ๋ฌธ์ด ๊ฒ€์ƒ‰๋˜์—ˆ์Šต๋‹ˆ๋‹ค.\n- ์†Œ์Šค๋ณ„ ๋ถ„ํฌ: pubmed(9), arxiv(1)\n- ์—ฐ๋„ ๋ฒ”์œ„: 2025-2025\n- ํ‰๊ท  ๋žญํ‚น ์ ์ˆ˜: 0.655\n\n์ฃผ์š” ๋ฐœ๊ฒฌ:\n- ์ƒ์œ„ ๋…ผ๋ฌธ๋“ค์€ ์ฃผ๋กœ Br J Oral Maxillofac Surg, World J Emerg Surg, Appl Clin Inform์—์„œ ๋ฐœํ‘œ๋˜์—ˆ์Šต๋‹ˆ๋‹ค.\n- ์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ: systematic_review(1), meta_analysis(1), survey(1), theoretical(5), experimental(2)", "charts": { "year_trend": "์—ฐ๋„๋ณ„ ํŠธ๋ Œ๋“œ ์ฐจํŠธ", "venue_distribution": "์ €๋„ ๋ถ„ํฌ ์ฐจํŠธ", "study_type_distribution": "์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ ์ฐจํŠธ" } }
What are the emerging topics in large language models?
10
๋‹ค์Œ ์งˆ๋ฌธ์— ๋Œ€ํ•ด ๋…ผ๋ฌธ์„ ๊ฒ€์ƒ‰ํ•˜๊ณ  ๋žญํ‚นํ•˜์—ฌ ๊ฒฐ๊ณผ๋ฅผ ์ œ์‹œํ•˜์„ธ์š”. ์งˆ๋ฌธ: What do systematic reviews say about attention mechanism? ์ถœ๋ ฅ ํ˜•์‹: 1. Evidence Table (Top-K): ๊ฒ€์ƒ‰๋œ ๋…ผ๋ฌธ์˜ ์ƒ์œ„ K๊ฐœ๋ฅผ ํ…Œ์ด๋ธ” ํ˜•์‹์œผ๋กœ ์ œ์‹œ 2. Ranking Breakdown: ๊ฐ ๋…ผ๋ฌธ์˜ ๋žญํ‚น ์ ์ˆ˜ ๊ตฌ์„ฑ ์š”์†Œ๋ฅผ ๊ณต๊ฐœ 3. Answer Summary: ๊ฒ€์ƒ‰ ๊ฒฐ๊ณผ๋ฅผ ์š”์•ฝํ•˜์—ฌ ์ œ์‹œ 4. Charts: ์—ฐ๋„๋ณ„ ํŠธ๋ Œ๋“œ, ์ €๋„ ๋ถ„ํฌ, ์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ ์ฐจํŠธ์— ๋Œ€ํ•œ ์„ค๋ช… ๋ชจ๋“  ์ฃผ์žฅ์€ Evidence Table์˜ ๋…ผ๋ฌธ์— ๊ทผ๊ฑฐํ•ด์•ผ ํ•˜๋ฉฐ, ๋žญํ‚น ๊ทผ๊ฑฐ๋ฅผ ๋ช…์‹œํ•ด์•ผ ํ•ฉ๋‹ˆ๋‹ค.
{ "evidence_table": [ { "rank": 1, "title": "What Do We Mean When We Talk about Trust in Social Media? A Systematic Review", "authors": "Yixuan Zhang, Joseph D Gaggiano, Nutchanon Yongsatianchot", "year": 2023, "venue": "arXiv", "study_type": "systematic_review", "score": 0.6425000000000001, "url": "https://arxiv.org/abs/2302.03671v1" }, { "rank": 2, "title": "Trust and Trustworthiness from Human-Centered Perspective in HRI -- A Systematic Literature Review", "authors": "Debora Firmino de Souza, Sonia Sousa, Kadri Kristjuhan-Ling", "year": 2025, "venue": "arXiv", "study_type": "systematic_review", "score": 0.6075, "url": "https://arxiv.org/abs/2501.19323v1" }, { "rank": 3, "title": "Computational-Assisted Systematic Review and Meta-Analysis (CASMA): Effect of a Subclass of GnRH-a on Endometriosis Recurrence", "authors": "Sandro Tsang", "year": 2025, "venue": "arXiv", "study_type": "meta_analysis", "score": 0.5975, "url": "https://arxiv.org/abs/2509.16599v3" }, { "rank": 4, "title": "Exploring the use of a Large Language Model for data extraction in systematic reviews: a rapid feasibility study", "authors": "Lena Schmidt, Kaitlyn Hair, Sergio Graziosi", "year": 2024, "venue": "arXiv", "study_type": "systematic_review", "score": 0.5825, "url": "https://arxiv.org/abs/2405.14445v2" }, { "rank": 5, "title": "Machine Learning Applications to Diffuse Reflectance Spectroscopy in Optical Diagnosis; A Systematic Review", "authors": "Nicola Rossberg, Celina L. Li, Simone Innocente", "year": 2025, "venue": "arXiv", "study_type": "systematic_review", "score": 0.5625, "url": "https://arxiv.org/abs/2503.02905v1" }, { "rank": 6, "title": "Non-invasive Techniques for Flow Rate Measurement in Water Pipes: Protocol for a Systematic Review", "authors": "Juan Diego Belesaca, Fabian Astudillo Salinas", "year": 2025, "venue": "arXiv", "study_type": "systematic_review", "score": 0.5625, "url": "https://arxiv.org/abs/2512.09239v1" }, { "rank": 7, "title": "Attention-Based Applications in Extended Reality to Support Autistic Users: A Systematic Review", "authors": "Katherine Wang, Simon Julier, Youngjun Cho", "year": 2022, "venue": "arXiv", "study_type": "systematic_review", "score": 0.5425, "url": "https://arxiv.org/abs/2204.00719v1" }, { "rank": 8, "title": "Pay Attention to What You Need", "authors": "Yifei Gao, Shaohong Chen, Lei Wang", "year": 2023, "venue": "arXiv", "study_type": "experimental", "score": 0.5225, "url": "https://arxiv.org/abs/2307.13365v3" }, { "rank": 9, "title": "A Systematic Review of Computational Thinking in Early Ages", "authors": "Edelberto Franco Silva, Bruno Josรฉ Dembogurski, Gustavo Silva Semaan", "year": 2021, "venue": "arXiv", "study_type": "systematic_review", "score": 0.49749999999999994, "url": "https://arxiv.org/abs/2106.10275v1" }, { "rank": 10, "title": "What Aspects of Mobile Ads Do Users Care About? An Empirical Study of Mobile In-app Ad Reviews", "authors": "Jiaping Gui, Meiyappan Nagappan, William G. J. Halfond", "year": 2017, "venue": "arXiv", "study_type": "experimental", "score": 0.48249999999999993, "url": "https://arxiv.org/abs/1702.07681v1" } ], "ranking_breakdown": [ { "paper_id": "2302.03671", "title": "What Do We Mean When We Talk about Trust in Social Media? A Systematic Review", "total_score": 0.6425000000000001, "query_match": 0.4, "recency": 0.8, "study_type_priority": 0.9, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2501.19323", "title": "Trust and Trustworthiness from Human-Centered Perspective in HRI -- A Systematic Literature Review", "total_score": 0.6075, "query_match": 0.2125, "recency": 1.0, "study_type_priority": 0.9, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2509.16599", "title": "Computational-Assisted Systematic Review and Meta-Analysis (CASMA): Effect of a Subclass of GnRH-a o", "total_score": 0.5975, "query_match": 0.1375, "recency": 1.0, "study_type_priority": 1.0, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2405.14445", "title": "Exploring the use of a Large Language Model for data extraction in systematic reviews: a rapid feasi", "total_score": 0.5825, "query_match": 0.2, "recency": 0.9, "study_type_priority": 0.9, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2503.02905", "title": "Machine Learning Applications to Diffuse Reflectance Spectroscopy in Optical Diagnosis; A Systematic", "total_score": 0.5625, "query_match": 0.1, "recency": 1.0, "study_type_priority": 0.9, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2512.09239", "title": "Non-invasive Techniques for Flow Rate Measurement in Water Pipes: Protocol for a Systematic Review", "total_score": 0.5625, "query_match": 0.1, "recency": 1.0, "study_type_priority": 0.9, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2204.00719", "title": "Attention-Based Applications in Extended Reality to Support Autistic Users: A Systematic Review", "total_score": 0.5425, "query_match": 0.2, "recency": 0.7, "study_type_priority": 0.9, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2307.13365", "title": "Pay Attention to What You Need", "total_score": 0.5225, "query_match": 0.2, "recency": 0.8, "study_type_priority": 0.7, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2106.10275", "title": "A Systematic Review of Computational Thinking in Early Ages", "total_score": 0.49749999999999994, "query_match": 0.1375, "recency": 0.6, "study_type_priority": 0.9, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "1702.07681", "title": "What Aspects of Mobile Ads Do Users Care About? An Empirical Study of Mobile In-app Ad Reviews", "total_score": 0.48249999999999993, "query_match": 0.4, "recency": 0.19999999999999996, "study_type_priority": 0.7, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 } ], "answer_summary": "๊ฒ€์ƒ‰ ์งˆ๋ฌธ: What do systematic reviews say about attention mechanism?\n\n๊ฒ€์ƒ‰ ๊ฒฐ๊ณผ ์š”์•ฝ:\n- ์ด 10๊ฐœ์˜ ๋…ผ๋ฌธ์ด ๊ฒ€์ƒ‰๋˜์—ˆ์Šต๋‹ˆ๋‹ค.\n- ์†Œ์Šค๋ณ„ ๋ถ„ํฌ: arxiv(10)\n- ์—ฐ๋„ ๋ฒ”์œ„: 2017-2025\n- ํ‰๊ท  ๋žญํ‚น ์ ์ˆ˜: 0.560\n\n์ฃผ์š” ๋ฐœ๊ฒฌ:\n- ์ƒ์œ„ ๋…ผ๋ฌธ๋“ค์€ ์ฃผ๋กœ arXiv, arXiv, arXiv์—์„œ ๋ฐœํ‘œ๋˜์—ˆ์Šต๋‹ˆ๋‹ค.\n- ์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ: systematic_review(7), meta_analysis(1), experimental(2)", "charts": { "year_trend": "์—ฐ๋„๋ณ„ ํŠธ๋ Œ๋“œ ์ฐจํŠธ", "venue_distribution": "์ €๋„ ๋ถ„ํฌ ์ฐจํŠธ", "study_type_distribution": "์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ ์ฐจํŠธ" } }
What do systematic reviews say about attention mechanism?
10
๋‹ค์Œ ์งˆ๋ฌธ์— ๋Œ€ํ•ด ๋…ผ๋ฌธ์„ ๊ฒ€์ƒ‰ํ•˜๊ณ  ๋žญํ‚นํ•˜์—ฌ ๊ฒฐ๊ณผ๋ฅผ ์ œ์‹œํ•˜์„ธ์š”. ์งˆ๋ฌธ: What are the emerging topics in reinforcement learning? ์ถœ๋ ฅ ํ˜•์‹: 1. Evidence Table (Top-K): ๊ฒ€์ƒ‰๋œ ๋…ผ๋ฌธ์˜ ์ƒ์œ„ K๊ฐœ๋ฅผ ํ…Œ์ด๋ธ” ํ˜•์‹์œผ๋กœ ์ œ์‹œ 2. Ranking Breakdown: ๊ฐ ๋…ผ๋ฌธ์˜ ๋žญํ‚น ์ ์ˆ˜ ๊ตฌ์„ฑ ์š”์†Œ๋ฅผ ๊ณต๊ฐœ 3. Answer Summary: ๊ฒ€์ƒ‰ ๊ฒฐ๊ณผ๋ฅผ ์š”์•ฝํ•˜์—ฌ ์ œ์‹œ 4. Charts: ์—ฐ๋„๋ณ„ ํŠธ๋ Œ๋“œ, ์ €๋„ ๋ถ„ํฌ, ์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ ์ฐจํŠธ์— ๋Œ€ํ•œ ์„ค๋ช… ๋ชจ๋“  ์ฃผ์žฅ์€ Evidence Table์˜ ๋…ผ๋ฌธ์— ๊ทผ๊ฑฐํ•ด์•ผ ํ•˜๋ฉฐ, ๋žญํ‚น ๊ทผ๊ฑฐ๋ฅผ ๋ช…์‹œํ•ด์•ผ ํ•ฉ๋‹ˆ๋‹ค.
{ "evidence_table": [ { "rank": 1, "title": "The future of robotic disassembly: a systematic review of techniques and applications in the age of AI.", "authors": "Ameur Soufiane, Tabaa Mohamed, Hidila Zineb", "year": 2025, "venue": "Front Robot AI", "study_type": "systematic_review", "score": 0.655, "url": "https://pubmed.ncbi.nlm.nih.gov/41140779/" }, { "rank": 2, "title": "Optimal Vasopressin Initiation in Septic Shock: The OVISS Reinforcement Learning Study.", "authors": "Kalimouttou Alexandre, Kennedy Jason N, Feng Jean", "year": 2025, "venue": "JAMA", "study_type": "theoretical", "score": 0.6299999999999999, "url": "https://pubmed.ncbi.nlm.nih.gov/40098600/" }, { "rank": 3, "title": "From literature to predictive modeling: Insights and machine learning applications from in vitro comet assays related to the genotoxicity of titanium dioxide nanomaterials.", "authors": "Furxhi Irini, Mirzaei Mahsa, Costa Anna", "year": 2025, "venue": "NanoImpact", "study_type": "experimental", "score": 0.6199999999999999, "url": "https://pubmed.ncbi.nlm.nih.gov/40273981/" }, { "rank": 4, "title": "Integrating habit science and learning theory to promote maintenance of behavior change: does adding text messages to a habit-based sleep health intervention (HABITs) improve outcomes for eveningness chronotype young adults? Study protocol for a randomized controlled trial.", "authors": "Diaz Marlen, Ovalle Patino Estephania, Oliver Sophia", "year": 2024, "venue": "Trials", "study_type": "randomized_trial", "score": 0.615, "url": "https://pubmed.ncbi.nlm.nih.gov/39563407/" }, { "rank": 5, "title": "COVID-19 and the domestication of adolescent and young people's sexuality in the emergence of the pandemic (2020-2022).", "authors": "Cabral Cristiane S, Paiva Vera, Barbosa Djalma", "year": 2025, "venue": "Cien Saude Colet", "study_type": "survey", "score": 0.6099999999999999, "url": "https://pubmed.ncbi.nlm.nih.gov/41337627/" }, { "rank": 6, "title": "Causal-Paced Deep Reinforcement Learning", "authors": "Geonwoo Cho, Jaegyun Im, Doyoon Kim", "year": 2025, "venue": "arXiv", "study_type": "experimental", "score": 0.6074999999999999, "url": "https://arxiv.org/abs/2507.02910v1" }, { "rank": 7, "title": "ATEM: A Topic Evolution Model for the Detection of Emerging Topics in Scientific Archives", "authors": "Hamed Rahimi, Hubert Naacke, Camelia Constantin", "year": 2023, "venue": "arXiv", "study_type": "experimental", "score": 0.6025, "url": "https://arxiv.org/abs/2306.02221v1" }, { "rank": 8, "title": "Simulation of human-vehicle interaction at right-turn unsignalized intersections: A game-theoretic deep maximum entropy inverse reinforcement learning method.", "authors": "Li Wenli, Li Xianglong, Li Lingxi", "year": 2025, "venue": "Accid Anal Prev", "study_type": "theoretical", "score": 0.595, "url": "https://pubmed.ncbi.nlm.nih.gov/39978040/" }, { "rank": 9, "title": "Designing, implementing and evaluating multidisciplinary healthcare training programmes in the wartime humanitarian context of Ukraine.", "authors": "Kivlehan Sean M, Niescierenko Michelle, Murray Kathleen", "year": 2025, "venue": "Emerg Med J", "study_type": "survey", "score": 0.595, "url": "https://pubmed.ncbi.nlm.nih.gov/39746801/" }, { "rank": 10, "title": "Machine-learning-assisted photonic device development: a multiscale approach from theory to characterization.", "authors": "Chen Yuheng, Montes McNeil Alexander, Park Taehyuk", "year": 2025, "venue": "Nanophotonics", "study_type": "experimental", "score": 0.59, "url": "https://pubmed.ncbi.nlm.nih.gov/41246485/" } ], "ranking_breakdown": [ { "paper_id": "41140779", "title": "The future of robotic disassembly: a systematic review of techniques and applications in the age of ", "total_score": 0.655, "query_match": 0.35, "recency": 1.0, "study_type_priority": 0.9, "evidence_completeness": 0.8999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "40098600", "title": "Optimal Vasopressin Initiation in Septic Shock: The OVISS Reinforcement Learning Study.", "total_score": 0.6299999999999999, "query_match": 0.45, "recency": 1.0, "study_type_priority": 0.5, "evidence_completeness": 0.9999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "40273981", "title": "From literature to predictive modeling: Insights and machine learning applications from in vitro com", "total_score": 0.6199999999999999, "query_match": 0.325, "recency": 1.0, "study_type_priority": 0.7, "evidence_completeness": 0.9999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "39563407", "title": "Integrating habit science and learning theory to promote maintenance of behavior change: does adding", "total_score": 0.615, "query_match": 0.3125, "recency": 0.9, "study_type_priority": 0.8, "evidence_completeness": 0.9999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "41337627", "title": "COVID-19 and the domestication of adolescent and young people's sexuality in the emergence of the pa", "total_score": 0.6099999999999999, "query_match": 0.35, "recency": 1.0, "study_type_priority": 0.6, "evidence_completeness": 0.9999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "2507.02910", "title": "Causal-Paced Deep Reinforcement Learning", "total_score": 0.6074999999999999, "query_match": 0.3125, "recency": 1.0, "study_type_priority": 0.7, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2306.02221", "title": "ATEM: A Topic Evolution Model for the Detection of Emerging Topics in Scientific Archives", "total_score": 0.6025, "query_match": 0.4, "recency": 0.8, "study_type_priority": 0.7, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "39978040", "title": "Simulation of human-vehicle interaction at right-turn unsignalized intersections: A game-theoretic d", "total_score": 0.595, "query_match": 0.3625, "recency": 1.0, "study_type_priority": 0.5, "evidence_completeness": 0.9999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "39746801", "title": "Designing, implementing and evaluating multidisciplinary healthcare training programmes in the warti", "total_score": 0.595, "query_match": 0.3125, "recency": 1.0, "study_type_priority": 0.6, "evidence_completeness": 0.9999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "41246485", "title": "Machine-learning-assisted photonic device development: a multiscale approach from theory to characte", "total_score": 0.59, "query_match": 0.2875, "recency": 1.0, "study_type_priority": 0.7, "evidence_completeness": 0.8999999999999999, "duplicate_penalty": 0.0 } ], "answer_summary": "๊ฒ€์ƒ‰ ์งˆ๋ฌธ: What are the emerging topics in reinforcement learning?\n\n๊ฒ€์ƒ‰ ๊ฒฐ๊ณผ ์š”์•ฝ:\n- ์ด 10๊ฐœ์˜ ๋…ผ๋ฌธ์ด ๊ฒ€์ƒ‰๋˜์—ˆ์Šต๋‹ˆ๋‹ค.\n- ์†Œ์Šค๋ณ„ ๋ถ„ํฌ: pubmed(8), arxiv(2)\n- ์—ฐ๋„ ๋ฒ”์œ„: 2023-2025\n- ํ‰๊ท  ๋žญํ‚น ์ ์ˆ˜: 0.612\n\n์ฃผ์š” ๋ฐœ๊ฒฌ:\n- ์ƒ์œ„ ๋…ผ๋ฌธ๋“ค์€ ์ฃผ๋กœ Front Robot AI, JAMA, NanoImpact์—์„œ ๋ฐœํ‘œ๋˜์—ˆ์Šต๋‹ˆ๋‹ค.\n- ์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ: systematic_review(1), theoretical(2), experimental(4), randomized_trial(1), survey(2)", "charts": { "year_trend": "์—ฐ๋„๋ณ„ ํŠธ๋ Œ๋“œ ์ฐจํŠธ", "venue_distribution": "์ €๋„ ๋ถ„ํฌ ์ฐจํŠธ", "study_type_distribution": "์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ ์ฐจํŠธ" } }
What are the emerging topics in reinforcement learning?
10
๋‹ค์Œ ์งˆ๋ฌธ์— ๋Œ€ํ•ด ๋…ผ๋ฌธ์„ ๊ฒ€์ƒ‰ํ•˜๊ณ  ๋žญํ‚นํ•˜์—ฌ ๊ฒฐ๊ณผ๋ฅผ ์ œ์‹œํ•˜์„ธ์š”. ์งˆ๋ฌธ: How has neural networks evolved over the past 5 years? ์ถœ๋ ฅ ํ˜•์‹: 1. Evidence Table (Top-K): ๊ฒ€์ƒ‰๋œ ๋…ผ๋ฌธ์˜ ์ƒ์œ„ K๊ฐœ๋ฅผ ํ…Œ์ด๋ธ” ํ˜•์‹์œผ๋กœ ์ œ์‹œ 2. Ranking Breakdown: ๊ฐ ๋…ผ๋ฌธ์˜ ๋žญํ‚น ์ ์ˆ˜ ๊ตฌ์„ฑ ์š”์†Œ๋ฅผ ๊ณต๊ฐœ 3. Answer Summary: ๊ฒ€์ƒ‰ ๊ฒฐ๊ณผ๋ฅผ ์š”์•ฝํ•˜์—ฌ ์ œ์‹œ 4. Charts: ์—ฐ๋„๋ณ„ ํŠธ๋ Œ๋“œ, ์ €๋„ ๋ถ„ํฌ, ์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ ์ฐจํŠธ์— ๋Œ€ํ•œ ์„ค๋ช… ๋ชจ๋“  ์ฃผ์žฅ์€ Evidence Table์˜ ๋…ผ๋ฌธ์— ๊ทผ๊ฑฐํ•ด์•ผ ํ•˜๋ฉฐ, ๋žญํ‚น ๊ทผ๊ฑฐ๋ฅผ ๋ช…์‹œํ•ด์•ผ ํ•ฉ๋‹ˆ๋‹ค.
{ "evidence_table": [ { "rank": 1, "title": "Review of applications of deep learning in veterinary diagnostics and animal health.", "authors": "Xiao Sam, Dhand Navneet K, Wang Zhiyong", "year": 2025, "venue": "Front Vet Sci", "study_type": "systematic_review", "score": 0.5870000000000001, "url": "https://pubmed.ncbi.nlm.nih.gov/40144529/" }, { "rank": 2, "title": "Multiclass Graph-Based Large Margin Classifiers: Unified Approach for Support Vectors and Neural Networks", "authors": "Vรญtor M. Hanriot, Luiz C. B. Torres, Antรดnio P. Braga", "year": 2025, "venue": "arXiv", "study_type": "experimental", "score": 0.5585, "url": "https://arxiv.org/abs/2512.13410v1" }, { "rank": 3, "title": "Tackling the Curse of Dimensionality with Physics-Informed Neural Networks", "authors": "Zheyuan Hu, Khemraj Shukla, George Em Karniadakis", "year": 2023, "venue": "arXiv", "study_type": "theoretical", "score": 0.5345, "url": "https://arxiv.org/abs/2307.12306v6" }, { "rank": 4, "title": "The Deep Arbitrary Polynomial Chaos Neural Network or how Deep Artificial Neural Networks could benefit from Data-Driven Homogeneous Chaos Theory", "authors": "Sergey Oladyshkin, Timothy Praditia, Ilja Krรถker", "year": 2023, "venue": "arXiv", "study_type": "theoretical", "score": 0.5185, "url": "https://arxiv.org/abs/2306.14753v1" }, { "rank": 5, "title": "Newton-Puiseux Analysis for Interpretability and Calibration of Complex-Valued Neural Networks", "authors": "Piotr Migus", "year": 2025, "venue": "arXiv", "study_type": "theoretical", "score": 0.5185, "url": "https://arxiv.org/abs/2504.19176v2" }, { "rank": 6, "title": "Learning Active Subspaces and Discovering Important Features with Gaussian Radial Basis Functions Neural Networks", "authors": "Danny D'Agostino, Ilija Ilievski, Christine Annette Shoemaker", "year": 2023, "venue": "arXiv", "study_type": "experimental", "score": 0.5065, "url": "https://arxiv.org/abs/2307.05639v2" }, { "rank": 7, "title": "Breast Cancer Detection and Diagnosis Using Mammographic Data: Systematic Review.", "authors": "Gardezi Syed Jamal Safdar, Elazab Ahmed, Lei Baiying", "year": 2019, "venue": "J Med Internet Res", "study_type": "systematic_review", "score": 0.506, "url": "https://pubmed.ncbi.nlm.nih.gov/31350843/" }, { "rank": 8, "title": "MECCH: Metapath Context Convolution-based Heterogeneous Graph Neural Networks", "authors": "Xinyu Fu, Irwin King", "year": 2022, "venue": "arXiv", "study_type": "experimental", "score": 0.49849999999999994, "url": "https://arxiv.org/abs/2211.12792v2" }, { "rank": 9, "title": "Dual Accuracy-Quality-Driven Neural Network for Prediction Interval Generation", "authors": "Giorgio Morales, John W. Sheppard", "year": 2022, "venue": "arXiv", "study_type": "experimental", "score": 0.49049999999999994, "url": "https://arxiv.org/abs/2212.06370v4" }, { "rank": 10, "title": "Continual Learning for Recurrent Neural Networks: an Empirical Evaluation", "authors": "Andrea Cossu, Antonio Carta, Vincenzo Lomonaco", "year": 2021, "venue": "arXiv", "study_type": "experimental", "score": 0.4784999999999999, "url": "https://arxiv.org/abs/2103.07492v4" } ], "ranking_breakdown": [ { "paper_id": "40144529", "title": "Review of applications of deep learning in veterinary diagnostics and animal health.", "total_score": 0.5870000000000001, "query_match": 0.18, "recency": 1.0, "study_type_priority": 0.9, "evidence_completeness": 0.8999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "2512.13410", "title": "Multiclass Graph-Based Large Margin Classifiers: Unified Approach for Support Vectors and Neural Net", "total_score": 0.5585, "query_match": 0.19, "recency": 1.0, "study_type_priority": 0.7, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2307.12306", "title": "Tackling the Curse of Dimensionality with Physics-Informed Neural Networks", "total_score": 0.5345, "query_match": 0.32999999999999996, "recency": 0.8, "study_type_priority": 0.5, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2306.14753", "title": "The Deep Arbitrary Polynomial Chaos Neural Network or how Deep Artificial Neural Networks could bene", "total_score": 0.5185, "query_match": 0.29000000000000004, "recency": 0.8, "study_type_priority": 0.5, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2504.19176", "title": "Newton-Puiseux Analysis for Interpretability and Calibration of Complex-Valued Neural Networks", "total_score": 0.5185, "query_match": 0.19, "recency": 1.0, "study_type_priority": 0.5, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2307.05639", "title": "Learning Active Subspaces and Discovering Important Features with Gaussian Radial Basis Functions Ne", "total_score": 0.5065, "query_match": 0.16, "recency": 0.8, "study_type_priority": 0.7, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "31350843", "title": "Breast Cancer Detection and Diagnosis Using Mammographic Data: Systematic Review.", "total_score": 0.506, "query_match": 0.24, "recency": 0.4, "study_type_priority": 0.9, "evidence_completeness": 0.9999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "2211.12792", "title": "MECCH: Metapath Context Convolution-based Heterogeneous Graph Neural Networks", "total_score": 0.49849999999999994, "query_match": 0.19, "recency": 0.7, "study_type_priority": 0.7, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2212.06370", "title": "Dual Accuracy-Quality-Driven Neural Network for Prediction Interval Generation", "total_score": 0.49049999999999994, "query_match": 0.16999999999999998, "recency": 0.7, "study_type_priority": 0.7, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "2103.07492", "title": "Continual Learning for Recurrent Neural Networks: an Empirical Evaluation", "total_score": 0.4784999999999999, "query_match": 0.19, "recency": 0.6, "study_type_priority": 0.7, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 } ], "answer_summary": "๊ฒ€์ƒ‰ ์งˆ๋ฌธ: How has neural networks evolved over the past 5 years?\n\n๊ฒ€์ƒ‰ ๊ฒฐ๊ณผ ์š”์•ฝ:\n- ์ด 10๊ฐœ์˜ ๋…ผ๋ฌธ์ด ๊ฒ€์ƒ‰๋˜์—ˆ์Šต๋‹ˆ๋‹ค.\n- ์†Œ์Šค๋ณ„ ๋ถ„ํฌ: pubmed(2), arxiv(8)\n- ์—ฐ๋„ ๋ฒ”์œ„: 2019-2025\n- ํ‰๊ท  ๋žญํ‚น ์ ์ˆ˜: 0.520\n\n์ฃผ์š” ๋ฐœ๊ฒฌ:\n- ์ƒ์œ„ ๋…ผ๋ฌธ๋“ค์€ ์ฃผ๋กœ Front Vet Sci, arXiv, arXiv์—์„œ ๋ฐœํ‘œ๋˜์—ˆ์Šต๋‹ˆ๋‹ค.\n- ์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ: systematic_review(2), experimental(5), theoretical(3)", "charts": { "year_trend": "์—ฐ๋„๋ณ„ ํŠธ๋ Œ๋“œ ์ฐจํŠธ", "venue_distribution": "์ €๋„ ๋ถ„ํฌ ์ฐจํŠธ", "study_type_distribution": "์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ ์ฐจํŠธ" } }
How has neural networks evolved over the past 5 years?
10
๋‹ค์Œ ์งˆ๋ฌธ์— ๋Œ€ํ•ด ๋…ผ๋ฌธ์„ ๊ฒ€์ƒ‰ํ•˜๊ณ  ๋žญํ‚นํ•˜์—ฌ ๊ฒฐ๊ณผ๋ฅผ ์ œ์‹œํ•˜์„ธ์š”. ์งˆ๋ฌธ: What are the differences between LSTM and RNN? ์ถœ๋ ฅ ํ˜•์‹: 1. Evidence Table (Top-K): ๊ฒ€์ƒ‰๋œ ๋…ผ๋ฌธ์˜ ์ƒ์œ„ K๊ฐœ๋ฅผ ํ…Œ์ด๋ธ” ํ˜•์‹์œผ๋กœ ์ œ์‹œ 2. Ranking Breakdown: ๊ฐ ๋…ผ๋ฌธ์˜ ๋žญํ‚น ์ ์ˆ˜ ๊ตฌ์„ฑ ์š”์†Œ๋ฅผ ๊ณต๊ฐœ 3. Answer Summary: ๊ฒ€์ƒ‰ ๊ฒฐ๊ณผ๋ฅผ ์š”์•ฝํ•˜์—ฌ ์ œ์‹œ 4. Charts: ์—ฐ๋„๋ณ„ ํŠธ๋ Œ๋“œ, ์ €๋„ ๋ถ„ํฌ, ์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ ์ฐจํŠธ์— ๋Œ€ํ•œ ์„ค๋ช… ๋ชจ๋“  ์ฃผ์žฅ์€ Evidence Table์˜ ๋…ผ๋ฌธ์— ๊ทผ๊ฑฐํ•ด์•ผ ํ•˜๋ฉฐ, ๋žญํ‚น ๊ทผ๊ฑฐ๋ฅผ ๋ช…์‹œํ•ด์•ผ ํ•ฉ๋‹ˆ๋‹ค.
{ "evidence_table": [ { "rank": 1, "title": "Carbon market price prediction in the Yangtze River Basin based on improved deep learning ensemble model with CEEMDAN and Attention-RNN.", "authors": "Lu Zuliang, Cao Zhihui, Xiang Zhuran", "year": 2025, "venue": "Carbon Balance Manag", "study_type": "experimental", "score": 0.625, "url": "https://pubmed.ncbi.nlm.nih.gov/41269474/" }, { "rank": 2, "title": "Developing and validating machine learning models to predict next-day extubation.", "authors": "Fenske Samuel W, Peltekian Alec, Kang Mengjia", "year": 2025, "venue": "Sci Rep", "study_type": "randomized_trial", "score": 0.6100000000000001, "url": "https://pubmed.ncbi.nlm.nih.gov/40731125/" }, { "rank": 3, "title": "Predicting Synergistic Drug Combinations Based on Fusion of Cell and Drug Molecular Structures.", "authors": "Yan Shiyu, Yu Gang, Yang Jiaoxing", "year": 2025, "venue": "Interdiscip Sci", "study_type": "experimental", "score": 0.605, "url": "https://pubmed.ncbi.nlm.nih.gov/40088336/" }, { "rank": 4, "title": "Sentiment analysis for deepfake X posts using novel transfer learning based word embedding and hybrid LGR approach.", "authors": "Khalid Madiha, Mushtaq Muhammad Faheem, Akram Urooj", "year": 2025, "venue": "Sci Rep", "study_type": "experimental", "score": 0.5900000000000001, "url": "https://pubmed.ncbi.nlm.nih.gov/40754634/" }, { "rank": 5, "title": "Performance analysis of neural network architectures for time series forecasting: A comparative study of RNN, LSTM, GRU, and hybrid models.", "authors": "Yunita Ariana, Pratama Mhd Iqbal, Almuzakki Muhammad Zaki", "year": 2025, "venue": "MethodsX", "study_type": "theoretical", "score": 0.5850000000000001, "url": "https://pubmed.ncbi.nlm.nih.gov/40777584/" }, { "rank": 6, "title": "A Long Short-Term Memory (LSTM) Model for Business Sentiment Analysis Based on Recurrent Neural Network", "authors": "Md. Jahidul Islam Razin, Md. Abdul Karim, M. F. Mridha", "year": 2025, "venue": "arXiv", "study_type": "experimental", "score": 0.5825, "url": "https://arxiv.org/abs/2509.03060v1" }, { "rank": 7, "title": "Development of student intent-based educational chatbot system with adaptive and attentive DTCN on symmetric convolution approach.", "authors": "Kathole Atul, Patil Suvarna, Jadhav Dr Devyani", "year": 2025, "venue": "MethodsX", "study_type": "experimental", "score": 0.5750000000000001, "url": "https://pubmed.ncbi.nlm.nih.gov/40822546/" }, { "rank": 8, "title": "Interdisciplinary framework for cyber-attacks and anomaly detection in industrial control systems using deep learning.", "authors": "Gulzar Qawsar, Mustafa Khurram", "year": 2025, "venue": "Sci Rep", "study_type": "experimental", "score": 0.5750000000000001, "url": "https://pubmed.ncbi.nlm.nih.gov/40695948/" }, { "rank": 9, "title": "Comparative analysis and enhancing rainfall prediction models for monthly rainfall prediction in the Eastern Thailand.", "authors": "Chuasuk Preeyanuch, Bhatrasataponkul Tachanat, Akkarapongtrakul Aniruj", "year": 2025, "venue": "MethodsX", "study_type": "theoretical", "score": 0.575, "url": "https://pubmed.ncbi.nlm.nih.gov/39802428/" }, { "rank": 10, "title": "Transfer learning for securing electric vehicle charging infrastructure from cyber-physical attacks.", "authors": "Almadhor Ahmad, Alsubai Shtwai, Bouazzi Imen", "year": 2025, "venue": "Sci Rep", "study_type": "experimental", "score": 0.5650000000000001, "url": "https://pubmed.ncbi.nlm.nih.gov/40102534/" } ], "ranking_breakdown": [ { "paper_id": "41269474", "title": "Carbon market price prediction in the Yangtze River Basin based on improved deep learning ensemble m", "total_score": 0.625, "query_match": 0.375, "recency": 1.0, "study_type_priority": 0.7, "evidence_completeness": 0.8999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "40731125", "title": "Developing and validating machine learning models to predict next-day extubation.", "total_score": 0.6100000000000001, "query_match": 0.25, "recency": 1.0, "study_type_priority": 0.8, "evidence_completeness": 0.9999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "40088336", "title": "Predicting Synergistic Drug Combinations Based on Fusion of Cell and Drug Molecular Structures.", "total_score": 0.605, "query_match": 0.2875, "recency": 1.0, "study_type_priority": 0.7, "evidence_completeness": 0.9999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "40754634", "title": "Sentiment analysis for deepfake X posts using novel transfer learning based word embedding and hybri", "total_score": 0.5900000000000001, "query_match": 0.25, "recency": 1.0, "study_type_priority": 0.7, "evidence_completeness": 0.9999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "40777584", "title": "Performance analysis of neural network architectures for time series forecasting: A comparative stud", "total_score": 0.5850000000000001, "query_match": 0.375, "recency": 1.0, "study_type_priority": 0.5, "evidence_completeness": 0.8999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "2509.03060", "title": "A Long Short-Term Memory (LSTM) Model for Business Sentiment Analysis Based on Recurrent Neural Netw", "total_score": 0.5825, "query_match": 0.25, "recency": 1.0, "study_type_priority": 0.7, "evidence_completeness": 0.95, "duplicate_penalty": 0.0 }, { "paper_id": "40822546", "title": "Development of student intent-based educational chatbot system with adaptive and attentive DTCN on s", "total_score": 0.5750000000000001, "query_match": 0.25, "recency": 1.0, "study_type_priority": 0.7, "evidence_completeness": 0.8999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "40695948", "title": "Interdisciplinary framework for cyber-attacks and anomaly detection in industrial control systems us", "total_score": 0.5750000000000001, "query_match": 0.25, "recency": 1.0, "study_type_priority": 0.7, "evidence_completeness": 0.8999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "39802428", "title": "Comparative analysis and enhancing rainfall prediction models for monthly rainfall prediction in the", "total_score": 0.575, "query_match": 0.35, "recency": 1.0, "study_type_priority": 0.5, "evidence_completeness": 0.8999999999999999, "duplicate_penalty": 0.0 }, { "paper_id": "40102534", "title": "Transfer learning for securing electric vehicle charging infrastructure from cyber-physical attacks.", "total_score": 0.5650000000000001, "query_match": 0.22499999999999998, "recency": 1.0, "study_type_priority": 0.7, "evidence_completeness": 0.8999999999999999, "duplicate_penalty": 0.0 } ], "answer_summary": "๊ฒ€์ƒ‰ ์งˆ๋ฌธ: What are the differences between LSTM and RNN?\n\n๊ฒ€์ƒ‰ ๊ฒฐ๊ณผ ์š”์•ฝ:\n- ์ด 10๊ฐœ์˜ ๋…ผ๋ฌธ์ด ๊ฒ€์ƒ‰๋˜์—ˆ์Šต๋‹ˆ๋‹ค.\n- ์†Œ์Šค๋ณ„ ๋ถ„ํฌ: pubmed(9), arxiv(1)\n- ์—ฐ๋„ ๋ฒ”์œ„: 2025-2025\n- ํ‰๊ท  ๋žญํ‚น ์ ์ˆ˜: 0.589\n\n์ฃผ์š” ๋ฐœ๊ฒฌ:\n- ์ƒ์œ„ ๋…ผ๋ฌธ๋“ค์€ ์ฃผ๋กœ Carbon Balance Manag, Sci Rep, Interdiscip Sci์—์„œ ๋ฐœํ‘œ๋˜์—ˆ์Šต๋‹ˆ๋‹ค.\n- ์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ: experimental(7), randomized_trial(1), theoretical(2)", "charts": { "year_trend": "์—ฐ๋„๋ณ„ ํŠธ๋ Œ๋“œ ์ฐจํŠธ", "venue_distribution": "์ €๋„ ๋ถ„ํฌ ์ฐจํŠธ", "study_type_distribution": "์—ฐ๊ตฌ ์œ ํ˜• ๋ถ„ํฌ ์ฐจํŠธ" } }
What are the differences between LSTM and RNN?
10
End of preview. Expand in Data Studio

No dataset card yet

Downloads last month
9