Improving Retrieval-Augmented Generation without Taxonomy-based Error Categorization

TL;DR

RePAIR enhances Retrieval-Augmented Generation by directly mapping flawed outputs to corrective actions without relying on detailed error categories, leading to improved performance across benchmarks.

Contribution

The paper introduces RePAIR, a novel approach that improves RAG systems without using explicit error taxonomies or critic supervision.

Findings

RePAIR consistently improves RAG performance on multiple benchmarks.

RePAIR eliminates the need for fine-grained error categorization.

RePAIR enhances robustness of error correction in RAG systems.

Abstract

Retrieval-Augmented Generation (RAG) improves the factual accuracy of large language model (LLM) outputs by grounding generation in external knowledge. Recent agentic RAG systems extend this paradigm with critical agents to evaluate model responses and iteratively refine outputs. However, most prior work implicitly assumes reliable critic feedback and focuses on planning strategies, while paying limited attention to the robustness of the error-correction process itself, which can be impacted by misaligned error categories and ineffective or incorrect corrections. Here, we hypothesize that RAG performance can be improved without explicit error categorization. We propose RePAIR, a response-action learning paradigm that directly maps flawed RAG outputs to error-mitigating action plans without relying on fine-grained error taxonomies and explicit critic supervision. Across multiple benchmarks, RePAIR consistently improves agentic RAG performance.