BRIEF: Bridging Retrieval and Inference for Multi-hop Reasoning via Compression
This addresses efficiency and performance issues in multi-hop question answering for users of retrieval-augmented LLMs, representing an incremental improvement over existing methods.
The paper tackles the problem of retrieval-augmented generation (RAG) for multi-hop reasoning, where increasing retrieved documents cause high latency and degraded performance, by introducing BRIEF, a method that compresses documents into dense summaries to improve efficiency and accuracy. It achieves a 2x compression rate improvement and outperforms state-of-the-art baselines by 3.00% EM and 4.16% F1 on HotpotQA.
Retrieval-augmented generation (RAG) can supplement large language models (LLMs) by integrating external knowledge. However, as the number of retrieved documents increases, the input length to LLMs grows linearly, causing a dramatic increase in latency and a degradation in long-context understanding. This is particularly serious for multi-hop questions that require a chain of reasoning across documents. To accelerate inference, reduce costs, and minimize distractions, this paper presents BRIEF (Bridging Retrieval and Inference through Evidence Fusion), a lightweight approach that performs query-aware multi-hop reasoning by compressing retrieved documents into highly dense textual summaries to integrate into in-context RAG. To enable learning compression for multi-hop reasoning, we curate synthetic data by extracting atomic propositions that encapsulate distinct factoids from the source documents to compose synthetic summaries. Based on our synthetic data built entirely by open-source models, BRIEF generates more concise summaries and enables a range of LLMs to achieve exceptional open-domain question answering (QA) performance. For example, on HotpotQA, BRIEF improves the compression rate by 2 times compared to the state-of-the-art baseline, while outperforming it by 3.00% EM and 4.16% F1 with Flan-UL2 as the reader model. It also generates more concise summaries than proprietary GPT-3.5, while demonstrating nearly identical QA performance.