STEPQuant: When and Where Errors Matter in Delta-Rule Recurrent State Quantization Paper • 2609.38169 • Published Sep 29 • 75
The Teacher Is a Direction, Not a Destination: Extrapolating RL-Induced Representation Residuals in On-Policy Distillation Paper • 2609.36484 • Published Sep 29 • 146
False Frontiers: Diagnosing and Mitigating Co-Cheating in Self-Evolving Search Agents Paper • 2609.39102 • Published Sep 30 • 130
YuE2: Unifying Symbolic and Audio Music Generation at Frontier Quality Paper • 2609.33757 • Published Sep 27 • 157
VisionHOPE: Visual Backbones as Self-Modifying Learning Systems Paper • 2609.33325 • Published Sep 27 • 53
LimiX-2: A Contextual Mechanism Network Towards General Structured-Data Intelligence Paper • 2609.17488 • Published Sep 15 • 235