Answering a complex question about a long video is a grounding problem: the answer turns on a few seconds of evidence, and vision-language models remain well below human accuracy at finding them. The standard way to spend extra test-time compute on this kind of video reasoning is best-of-N selection: sample several cha...
Martin Q. Ma, Yu-Xiao Qu, Ruslan Salakhutdinov et al.· 0 citations
Foundation models are typically trained in stages: pretraining, midtraining, supervised fine-tuning (SFT), and reinforcement learning (RL), with changes in both data and learning algorithms. We show that this algorithmic separation is not fundamental. By deriving suitable RL objectives as maximum-likelihood fitting to...
Yu-Da Song, Alvin Zhang, Fahim Tajwar et al.· 0 citations
This work proposes training models to fail gracefully: to preserve successful behavior whenever possible while learning to execute a safe fallback policy when acting would be unreliable, and shows that reinforcement learning can not only incentivize models to succeed, but also to reason about the limits of their capabi...
Daman Arora, Xue-Chen Liu, H. Singh et al.· 0 citations
Meta-reasoning, an inference-time harness that makes agentic meta-reasoning an explicit and structured reasoning process, keeps improving over the tested budget ranges where direct control plateaus, though its overhead can hurt at small budgets.
P. Dahal, A. Bakhtin, Taco Cohen et al.· arXiv.org· 1 citation
This work introduces harness learning, which trains a proposer model to revise a solver's harness using execution feedback, and formulate this process as meta-learning over executable programs, with harness revisions playing the role of weight updates in gradient-based adaptation.
Alvin Zhang, Xue-Chen Liu, Zi-Xuan Wang et al.· arXiv.org· 0 citations
Tail-Likelihood Reinforcement Learning (TailRL), which maximizes the log-probability of exceeding a randomly chosen reward threshold, which gives more weight to rare, high-reward rollouts and can be interpreted as a mixture of Best-of-k gradients.
Shrinivas Ramasubramanian, Daman Arora, Fahim Tajwar et al.· 2 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.