This paper introduces On-policy Power Distillation (OPPD), a method that sharpens a language model's output probabilities to improve reasoning without changing parameters. By training the model to generate answers in one step, OPPD achieves significant accuracy gains on math and coding benchmarks compared to traditional sampling methods and other techniques like GRPO.
The chart will appear after repeat observations. The current metric comes from the source.
0 source votesReal observations only. History before source connection is not reconstructed.
This method improves reasoning accuracy by sharpening probability distributions without requiring additional computation during inference, making it efficient for real-time applications.
The paper addresses co-cheating in self-evolving search agents, where proposers and solvers develop shared errors leading to misleading internal rewards. It introduces Multi-Sample Verification (MSV) and CrossFit methods to mitigate this issue, showing improvements in reducing false agreement and enhancing search performance.
6 days agoDaily PapersRealCompanion introduces a benchmark for evaluating AI's ability to understand humans through long-term conversations, featuring 27,218 messages from 10 real relationships. The study reveals that most queries don't require long-term memory and that current methods struggle to identify when memory is needed.
5 days agoDaily PapersThis survey explores post-training and alignment techniques in video generation models, highlighting challenges like temporal coherence and physical constraints. It categorizes methods into supervised fine-tuning, self-training, preference-based approaches, and inference-time strategies, while discussing datasets, evaluation practices, and open challenges.
6 days agoDaily PapersEgo2Act is a benchmark for evaluating goal-directed manipulation in egocentric video generation, featuring 2,640 videos from 110 real-world tasks. It assesses whether video generation models can create realistic egocentric videos of a hand performing multi-step object manipulations to achieve high-level goals.
5 days agoDaily PapersThis paper introduces ProWAM, a progressive world action model that predicts actions and sparse visual sub-goals to guide robotic control. It improves long-horizon planning by using a single video-backbone pass for efficient action generation and demonstrates strong performance on simulation and real-world benchmarks.
5 days agoLessWrongThe text discusses a scenario where an AI must respond to the question 'What's the date?' without access to real-time data. It explores the challenge of providing an accurate date without hallucinating, considering OpenAI's training focus on avoiding false information. The text speculates on potential model versions and knowledge cutoff dates but acknowledges uncertainty due to lack of explicit information.
5 days ago