Effective Large Language Model Adaptation for Improved Grounding

Anonymous

Abstract: 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. We start with the design of a test-time adaptation capability that can retrieve passages to support the claims that have not been grounded, which iteratively improves the responses of LLMs. To effectively enable this capability, we propose tuning LLMs to self-ground the claims in their responses and provide accurate citations. 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. Across five datasets and two LLMs, results show that our tuning-based AGREE framework generates superior grounded responses with more accurate citations compared to prompting-based approaches and post-hoc citing-based approaches.

Paper Type: long

Research Area: Question Answering

Languages Studied: English