nerkyor/Qwen3.8-27B-Coder390-EfficientThink-Opus5.5-GPT6Astra-Grok4.7-DSV4Pro-K3-SFT-RLOO-MTP-DFlash2 Image-Text-to-Text • 27B • Updated 4 days ago • 365k • 472
Continually Post-training Small Models via Synthetic Self-distillation Paper • 2610.13113 • Published 14 days ago • 47
SPW-Nav: A Streaming Panoramic World Model for Language-Guided Navigation Paper • 2610.08941 • Published 17 days ago • 6
OSWorld-Pro: Process-based Evaluation for Computer Use Agents Paper • 2609.24890 • Published Sep 21 • 28
Equal Ranking Quality, Different Decisions: Measuring and Reducing Order Dependence in LLM Scorers Paper • 2608.26762 • Published 27 days ago • 24
OTRetarget: Joint Robot and Object Motion Retargeting via Optimal Transport Paper • 2609.36602 • Published 24 days ago • 8
EvoSafeHarness: Evolving Model- and Domain-Specific Harnesses for Securing Agents Paper • 2609.05903 • Published Sep 5 • 64
Beyond Retrieval: Progressive Latent Memory Evolution for Streaming Video Understanding Paper • 2609.04131 • Published Sep 3 • 31