The Teacher Is a Direction, Not a Destination: Extrapolating RL-Induced Representation Residuals in On-Policy Distillation Paper • 2609.36484 • Published 17 days ago • 562
False Frontiers: Diagnosing and Mitigating Co-Cheating in Self-Evolving Search Agents Paper • 2609.39102 • Published 16 days ago • 664
Transformers Stop Thinking Too Early, and a Tiny LoRA Fixes It Paper • 2609.36585 • Published 17 days ago • 72
NCP-ArchPreview Technical Report: Moving towards Latent Space Language Models through Next Concept Prediction Paper • 2609.10715 • Published Sep 9 • 269