面向改进的接地与引用生成的高效大语言模型适配
Effective Large Language Model Adaptation for Improved Grounding and Citation Generation
AI总结：
本文提出 AGREE 调优框架，通过自动构建含引用的接地数据并结合测试时主动检索，提升大语言模型回答的事实接地与引用准确性。
摘要：
大语言模型（LLM）在自然语言理解与生成方面取得了显著进展。然而，阻碍其在现实世界中广泛部署的一个主要问题是，它们可能生成不符合事实的“幻觉”答案。为此，本文聚焦于通过将 LLM 的回答接地于检索段落并提供引用来改进 LLM。我们提出了一个新框架 AGREE，即 Adaptation for GRounding EnhancEment（用于接地增强的适配），从整体视角改进接地能力。该框架对 LLM 进行调优，使其对回答中的主张进行自我接地，并为检索到的文档提供准确引用。在预训练 LLM 之上进行这种调优，需要针对成对查询具备良好接地的回答（含引用），为此我们引入了一种能够从未标注查询自动构建此类数据的方法。经过调优的 LLM 的自我接地能力进一步赋予其测试时适配（TTA）能力：能够主动检索段落以支持尚未接地的主张，从而迭代式改进 LLM 的回答。在五个数据集和两个 LLM 上的结果表明，与基于提示的方法和基于事后引用的方法相比，所提出的基于调优的 AGREE 框架能够生成更优的接地回答，并具有更准确的引用。
英文摘要：
Large language models (LLMs) have achieved remarkable advancements in natural language understanding and generation. However, one major issue towards their widespread deployment in the real world is that they can generate "hallucinated" answers that are not factual. Towards this end, this paper focuses on improving LLMs by grounding their responses in retrieved passages and by providing citations. We propose a new framework, AGREE, Adaptation for GRounding EnhancEment, that improves the grounding from a holistic perspective. Our framework tunes LLMs to selfground the claims in their responses and provide accurate citations to retrieved documents. This tuning on top of the pre-trained LLMs requires well-grounded responses (with citations) for paired queries, for which we introduce a method that can automatically construct such data from unlabeled queries. The selfgrounding capability of tuned LLMs further grants them a test-time adaptation (TTA) capability that can actively retrieve passages to support the claims that have not been grounded, which iteratively improves the responses of LLMs. Across five datasets and two LLMs, our results show that the proposed tuningbased AGREE framework generates superior grounded responses with more accurate citations compared to prompting-based approaches and post-hoc citing-based approaches
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。
