nerkyor/Qwen3.8-27B-Coder390-EfficientThink-Opus5.5-GPT6Astra-Grok4.7-DSV4Pro-K3-SFT-RLOO-MTP-DFlash2 Image-Text-to-Text • 27B • Updated 1 day ago • 75.3k • 292
Accurate but Not Humble: Evaluating Epistemic Humility in LLM Agents under Knowledge Conflict Paper • 2610.12360 • Published 5 days ago • 10
When EOS Tokens Disagree: Understanding Length Inflation in On-Policy Distillation Paper • 2609.20511 • Published 26 days ago • 115
Continual Learning Mechanisms Compose for Long-Horizon Memorization Paper • 2609.06986 • Published Sep 7 • 376