Peter Szemraj PRO

pszemraj's activity

• All
• Models
• Datasets
• Spaces
• Buckets
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

upvoted a paper about 23 hours ago

RelateAnything: Real-Time Open-Vocabulary Relation Prediction From Any Inputs

Paper • 2609.12552 • Published 7 days ago • 3

upvoted a paper 2 days ago

How Lossless Is Lossless Speculative Decoding? The Role of Numerical Precision in Orthrus

Paper • 2609.15504 • Published 4 days ago • 30

upvoted an article 3 days ago

view article

Article

IBM releases SOTA Granite Time Series PatchTST-FM-r2 model with commercial-friendly license

• 

ibm-research

•

8 days ago

• 58

upvoted a paper 5 days ago

MetroLLM-Bench: Evaluating Language Models as Transit Kiosk Runtimes

Paper • 2609.10016 • Published 9 days ago • 33

upvoted a paper 6 days ago

NCP-ArchPreview Technical Report: Moving towards Latent Space Language Models through Next Concept Prediction

Paper • 2609.10715 • Published 9 days ago • 320

upvoted a paper 10 days ago

Don't Drop Dropout: Optimizing Layer Sparsity for Efficient LLM Training and Inference

Paper • 2609.05275 • Published 14 days ago • 23

upvoted a collection 11 days ago

VibeVoice

Collection

Frontier Text-to-Speech Models https://microsoft.github.io/VibeVoice/ • 11 items • Updated 15 days ago • 263

upvoted an article 12 days ago

view article

Article

NeoMME: an efficient Multimodal-native and Multilingual Encoder

• 

Hcompany

•

14 days ago

• 104

upvoted a collection 14 days ago

Institutional Newspapers

Collection

A growing corpus of newspapers, parsed and optimized for computational access. • 6 items • Updated 3 days ago • 8

upvoted a paper 17 days ago

Puro-2B: Poor Lab's Qwen2-1.5B Trained on RTX 5090 within $5090

Paper • 2608.27370 • Published 22 days ago • 41

upvoted a paper 26 days ago

The Embedder's Dilemma: LLMs Are Better, but at What Cost?

Paper • 2608.12875 • Published Aug 13 • 15

upvoted a paper 27 days ago

Thinking in a Low-Resource Language: What SFT Builds, What RL Fixes, What Accuracy Cannot See

Paper • 2608.17744 • Published about 1 month ago • 16

upvoted a paper 28 days ago

FM-Bench: A Benchmark for Long-Horizon Management with Competing Agents

Paper • 2608.18423 • Published 30 days ago • 21

upvoted 2 papers about 1 month ago

Scaling Domain Data Repetition in LLM Pretraining

Paper • 2608.14071 • Published Aug 14 • 16

SimpleOPD: Simple Tokenizer-Agnostic On-Policy Distillation for Long-Context Reasoning

Paper • 2608.14277 • Published Aug 14 • 36

upvoted an article about 1 month ago

view article

Article

Meet North Micro Vision: A 2.4B Native-Resolution Vision-Language Model

• 

CohereLabs

•

Aug 12

• 40

upvoted 4 papers about 1 month ago

Full-bandwidth transformer

Paper • 2608.08888 • Published Aug 9 • 24

Massive Activations in Hybrid Linear Attention Large Language Models: Pre-Attention Spikes and Inter-Spike Plateaus

Paper • 2608.12149 • Published Aug 12 • 30

LLMRouter: Unified Infrastructure for Developing, Evaluating, and Deploying LLM Routers

Paper • 2608.06867 • Published Aug 7 • 113

Poor Man's Agentic Modeling: Simulating Large LLM-Agent Societies on a Laptop

Paper • 2608.11215 • Published Jul 19 • 7