Fine-Grained Self-Endorsement Improves Factuality and Reasoning

TL;DR

This paper introduces a self-endorsement method that compares fact-level details across multiple responses to reduce hallucinations and improve factual accuracy in large language model outputs, especially for long-form tasks.

Contribution

The proposed approach leverages fine-grained fact comparisons for response selection, outperforming prior ensemble methods in mitigating hallucinations in LLMs.

Findings

Effective in improving factuality across different LLM scales

Works well with simple prompts for longform generation

Shows potential for broader applications in factual consistency

Abstract

This work studies improving large language model (LLM) generations at inference time by mitigating fact-conflicting hallucinations. Particularly, we propose a self-endorsement framework that leverages the fine-grained fact-level comparisons across multiple sampled responses. Compared with prior ensemble methods (Wang et al., 2022;Chen et al., 2023)) that perform response-level selection, our approach can better alleviate hallucinations, especially for longform generation tasks. Our approach can broadly benefit smaller and open-source LLMs as it mainly conducts simple content-based comparisons. Experiments on Biographies show that our method can effectively improve the factuality of generations with simple and intuitive prompts across different scales of LLMs. Besides, comprehensive analyses on TriviaQA and GSM8K demonstrate the potential of self-endorsement for broader application.