Improving Retrieval-Augmented Generation without Taxonomy-based Error Categorization

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhang, Gongbo, Peng, Yifan, Weng, Chunhua
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911695910207488
author Zhang, Gongbo
Peng, Yifan
Weng, Chunhua
author_facet Zhang, Gongbo
Peng, Yifan
Weng, Chunhua
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2605_18772
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Improving Retrieval-Augmented Generation without Taxonomy-based Error Categorization
Zhang, Gongbo
Peng, Yifan
Weng, Chunhua
Information Retrieval
Artificial Intelligence
Computation and Language
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.
title Improving Retrieval-Augmented Generation without Taxonomy-based Error Categorization
topic Information Retrieval
Artificial Intelligence
Computation and Language
url https://arxiv.org/abs/2605.18772