Papers
Deep, long-form breakdowns of recent research in AI, machine learning, security & programming languages.
Stellar Colosseum: A Many-Agent Harness for Long-Horizon Research in Mathematics and Theoretical Computer Science
by Honghao Lin, David P. Woodruff, Yuan Deng, Jieming Mao, Song Zuo, Vahab Mirrokni
Disentangling Representation Evolution in Transformers through Directional Decomposition
by Shwai He, Haichao Zhang, Shen Yan
ResSafe: Learning Safety Filtering with Residual Reinforcement Learning for Humanoids
by Gechen Qu, Tong Zhang, Bike Zhang, Yen-Jen Wang, Koushil Sreenath, Claire Tomlin, Jason Jangho Choi
Bellman Policy Optimization
by Zhuoqing Song, Haotian Xu, Xikun Zhang, Lidong Bing
The Router Within: Eliciting Native Skill Routing from a Frozen LLM
by Ruishuo Chen, Xun Wang, Yu Chen, Zhuoran Li, Longbo Huang
A Chosen Future Can Still Be Rewritten: Causal Writability in Video Models
by Xingyun Wang, Haomin Zheng, Man Yuan, Leqian Yang, Ziming Liu
Discovery Foundation Models: Toward Open-Ended Discovery Intelligence
by Ling Yang, Zhenfei Yin, Yingcheng Wu
Verifiable by Construction: Claim-Level Evaluation of Verbatim Citation in Clinical Question Answering
by Jiashuo Zhang, Yuling Chen, Yvonne Commodore-Mensah, Michael Oberst
SNAP3D: Physically Grounded 3D Parts for Assembly from a Single Image
by Yu-Rou Tuan, Hao-Tang Tsui, Nicolas Ugrinovic, Kris Kitani, Xiaoxuan Ma
Type Diversity Enables Transformers to Generalise Compositionally
by Anssi Moisio, Mathias Creutz, Mikko Kurimo