Bridging Relevance and Reasoning: Rationale Distillation in Retrieval-Augmented Generation

ACL ARR 2024 December Submission43 Authors

Abstract: The reranker and generator are two critical components in the Retrieval-Augmented Generation (i.e., RAG) pipeline, responsible for ranking relevant documents and generating responses. However, due to differences in pre-training data and objectives, there is an inevitable gap between the documents ranked as relevant by the reranker and those required by the generator to support answering the query. To address this gap, we propose **RADIO**, a novel and practical preference alignment framework with **RA**tionale **DI**stillati**O**n. Specifically, We first propose a rationale extraction method that leverages the reasoning capabilities of large language models (LLMs) to extract the rationales necessary for answering the query. Subsequently, a rationale-based alignment process is designed to rerank the documents based on the extracted rationales, and fine-tune the reranker to align the preferences. We conduct extensive experiments on two tasks across three datasets to demonstrate the effectiveness of our approach compared to baseline methods. Our code is released online to ease reproduction.

Paper Type: Long

Research Area: Generation

Research Area Keywords: retrieval-augmented generation, re-ranking

Contribution Types: Model analysis & interpretability, NLP engineering experiment, Publicly available software and/or pre-trained models

Languages Studied: English

Submission Number: 43