hill-sampling

标签

Cards List
#hill-sampling

Hill Sampling for Test-Time Scaling: A Simpler, Superior Alternative to Repeated Sampling, Evolution, and Training

arXiv cs.LG · 昨天 缓存

Hill Sampling is a simple test-time scaling method that uses frozen LLMs to iteratively sample and edit the best-verified program, achieving state-of-the-art results on algorithmic problems like circle packing and Erdős' minimum-overlap problem.

0 人收藏 0 人点赞
← 返回首页

提交意见反馈