Select-And-Extract: A Lightweight Plugin for Retrieval-Augmented Generation

2026-08-01

AI Summary

This work addresses retrieval and reading failures in Retrieval-Augmented Generation (RAG) by introducing SANE, a lightweight plug-in that enhances RAG efficiency without substantially increasing computational overhead. SANE first retrieves candidate documents through semantic search, then employs a language model–driven summarization step to filter key content. It further incorporates a blueprint-guided, query-time evidence extraction mechanism to supply the generator with concise, structured reasoning support. By avoiding complex architectures or excessive language model invocations, SANE simultaneously mitigates both retrieval and reading failures. Extensive evaluations across multiple benchmark tasks demonstrate significant performance gains, confirming the effectiveness of this lightweight design for efficient RAG systems.

Abstract

Retrieval-augmented generation (RAG) for language model (LM) systems fundamentally has two failure modes: retrieval failure and reading failure. The former fails to recall the right pieces of information from the external corpus, and the latter fails to produce the correct answer although the right information is retrieved. Some methods perform structured indexing for retrieval failure, but may suffer from limited generalization of the fixed structures. Some methods perform query-time structuring for reading failure, but typically require a lot of LM calls and rely heavily on the LM's capability. To this end, we propose Select-ANd-Extract (SANE), a simple yet effective plugin for RAG. For the retrieval failure, we retrieve a wide set of candidates with a semantic retriever, and leverage the LM to select the top candidates based on their synopses, which yields better recall than the original retriever. For the reading failure, we perform blueprint-guided query-time evidence extraction, which allows the generator LM to use only compact and structured key information so that it can perform better reasoning. Empirical results confirm that SANE brings solid improvements, while only introducing modest extra overhead. As a lightweight plugin for RAG, SANE offers a simple alternative to heavier approaches, and suggests a high-performance RAG framework need not be overly complex.

Problem

• retrieval-augmented generation
• retrieval failure
• reading failure
• language models
• information retrieval

Innovation

• Retrieval-Augmented Generation
• Select-And-Extract
• Evidence Extraction
• Semantic Retrieval
• Lightweight Plugin