Submitted by schlegelm 62 Emergent temporal abstractions in autoregressive models enable hierarchical reinforcement learning Google 5
Submitted by Jinjing713 33 Spatia: Video Generation with Updatable Spatial Memory The University of Sydney 167 4
Submitted by zhoutianyi 16 Schoenfeld's Anatomy of Mathematical Reasoning by Language Models · 5 authors 21 5
Submitted by Xiaoye08 12 VA-π: Variational Policy Alignment for Pixel-Aware Autoregressive Generation · 7 authors 12 3
Submitted by zxhuang1698 10 How Much 3D Do Video Foundation Models Encode? University of Illinois at Urbana-Champaign 3
Submitted by weit123 5 GTR-Turbo: Merged Checkpoint is Secretly a Free Teacher for Agentic VLM Training · 7 authors 3