Soft Self-Consistency Improves Language Model Agents

arXiv:2402.13212v1 Announce Type: cross
Abstract: Generations from large language models (LLMs) can be improved by sampling and scoring multiple solutions to select a final answer. Current "sample and select" methods such as self-consistency (SC) rely on majority voting to score answers. However, when tasks have many distinct and valid answers, selection by voting requires a large number of samples. This makes SC prohibitively expensive for interactive tasks that involve generating multiple actions (answers) sequentially. After establishing that majority voting fails …

abstract agents arxiv cs.ai cs.cl cs.lg current language language model language models large language large language models llms multiple sample sampling scoring solutions tasks type voting