The Teacher Is a Direction, Not a Destination: Extrapolating RL-Induced Representation Residuals in On-Policy Distillation Paper • 2609.36484 • Published 7 days ago • 577
MaLiang-Harness: A Programmable Path to Image and Video Generation Paper • 2609.34309 • Published 8 days ago • 412
VoxMem: Benchmarking Multimodal Memory in Large Audio Language Models Paper • 2609.32607 • Published 10 days ago • 154
Vidu S2: Real-Time Interactive, Editable, and Spatial Video Generation Paper • 2609.11638 • Published 26 days ago • 707
NeoHorse-1: Towards Recursive Self-Improvement via Agentic Post-Training with Routing Harness Paper • 2609.08183 • Published 28 days ago • 327