AI systems now write long and difficult mathematical proofs, and checking their step-by-step validity is becoming a bottleneck. A learned verifier could check proofs cheaply, but training one requires a correctness label on every step, and such labels are hard to obtain for open-ended proofs. Existing methods take them...
Rodney Lafuente-Mercado, Salman Rahman, Jing-Yan Shen et al.· 0 citations
By making failure cheap to locate and correct, this work is a foundation for more trustworthy long-horizon agents that learn from their own mistakes, and a practical path to overseeing increasingly autonomous AI.
Salman Rahman, Y. Kim, Mihir Parmar et al.· 2 citations
DiscoverPhysics, an interactive benchmark that asks a LLM agent to discover the laws of motion of a simulated world whose physics deliberately deviates from the authors' own, is introduced.
M. Wiemann, Lindsay M. Smith, Peter Melchior et al.· arXiv.org· 12 citations· ⚡2
A quantitative account of the pretraining-to-RL interface and a controlled testbed for studying the science of reasoning across the full pretraining-to-post-training pipeline is provided.
Jingyan Shen, Ang Li, Salman Rahman et al.· 0 citations
This work introduces Code Monitor Red Teaming, a monitor-red-teaming protocol that fixes a public-check information boundary while varying generator pressure, verifier scaffolding, and weak-to-strong capability.
Jun-Hui Liao, Jiawen Deng, Fuji Ren et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
