LLMs是偏向评估者但不偏向检索增强生成
LLMs are Biased Evaluators But Not Biased for Retrieval Augmented Generation
AI总结：
研究发现LLM在RAG框架中无自我偏好，但事实准确性显著影响输出。
摘要：
近期研究表明，大型语言模型（LLMs）在评估任务中表现出显著的偏见，特别是在倾向于评分和偏爱自动生成内容方面。然而，这种偏见在事实性任务中的表现，尤其是在检索增强生成（RAG）框架中，其中关键词提取和事实准确性优先于风格元素，仍然不明确。我们的研究通过模拟RAG框架的两个关键阶段来解决这一知识空白。第一阶段，LLMs评估人类撰写和模型生成的段落，模拟了pointwise reranking阶段。第二阶段涉及进行成对阅读理解测试以模拟生成阶段。与之前发现的评分任务中的自我偏好相反，我们的结果表明在RAG框架中没有显著的自我偏好效应。相反，我们发现事实准确性显著影响LLMs的输出，即使在没有先验知识的情况下。这些发现一致出现在三个常见的问答数据集（NQ、MARCO、TriviaQA数据集）和五个广泛采用的语言模型（GPT-3.5、GPT-4o-mini、Gemini、LLaMA3和Mistral）中。我们的研究为LLM偏见及其对基于RAG系统的含义的持续讨论做出了贡献，提供了可能指导开发更稳健和无偏的LLM系统的见解。
英文摘要：
Recent studies have demonstrated that large language models (LLMs) exhibit significant biases in evaluation tasks, particularly in preferentially rating and favoring self-generated content. However, the extent to which this bias manifests in fact-oriented tasks, especially within retrieval-augmented generation (RAG) frameworks, where keyword extraction and factual accuracy take precedence over stylistic elements, remains unclear. Our study addresses this knowledge gap by simulating two critical phases of the RAG framework. In the first phase, LLMs evaluated human-authored and model-generated passages, emulating the \textit{pointwise reranking phase}. The second phase involves conducting pairwise reading comprehension tests to simulate the \textit{generation phase}. Contrary to previous findings indicating a self-preference in rating tasks, our results reveal no significant self-preference effect in RAG frameworks. Instead, we observe that factual accuracy significantly influences LLMs' output, even in the absence of prior knowledge. These findings are consistent among three common QA datasets (NQ, MARCO, TriviaQA Datasets) and 5 widely adopted language models (GPT-3.5, GPT-4o-mini, Gemini, LLaMA3, and Mistral). Our research contributes to the ongoing discourse on LLM biases and their implications for RAG-based system, offering insights that may inform the development of more robust and unbiased LLM systems.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
