Related papers: RAGXplain: From Explainable Evaluation to Actionable Guidance of RAG Pipelines

RAGXplain: From Explainable Evaluation to Actionable Guidance of RAG Pipelines

URL: http://arxiv.org/abs/2505.13538v1
Date: Sun, 18 May 2025 17:25:34 GMT
Title: RAGXplain: From Explainable Evaluation to Actionable Guidance of RAG Pipelines
Authors: Dvir Cohen, Lin Burg, Gilad Barkan,
Abstract summary: RAGXplain is an evaluation framework that quantifies RAG performance and translates these assessments into clear insights.<n> RAGXplain thus bridges quantitative evaluation and practical optimization, empowering users to understand, trust, and enhance their AI systems.
Score: 0.7373617024876725
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Abstract: Retrieval-Augmented Generation (RAG) systems show promise by coupling large language models with external knowledge, yet traditional RAG evaluation methods primarily report quantitative scores while offering limited actionable guidance for refining these complex pipelines. In this paper, we introduce RAGXplain, an evaluation framework that quantifies RAG performance and translates these assessments into clear insights that clarify the workings of its complex, multi-stage pipeline and offer actionable recommendations. Using LLM reasoning, RAGXplain converts raw scores into coherent narratives identifying performance gaps and suggesting targeted improvements. By providing transparent explanations for AI decision-making, our framework fosters user trust-a key challenge in AI adoption. Our LLM-based metric assessments show strong alignment with human judgments, and experiments on public question-answering datasets confirm that applying RAGXplain's actionable recommendations measurably improves system performance. RAGXplain thus bridges quantitative evaluation and practical optimization, empowering users to understand, trust, and enhance their AI systems.

Related papers

FIRE: Faithful Interpretable Recommendation Explanations [2.6499018693213316]
Natural language explanations in recommender systems are often framed as a review generation task.<n>Fire is a lightweight and interpretable framework that combines SHAP-based feature attribution with structured, prompt-driven language generation.<n>Our results demonstrate that FIRE not only achieves competitive recommendation accuracy but also significantly improves explanation quality along critical dimensions such as alignment, structure, and faithfulness.
arXiv Detail & Related papers (2025-08-07T10:11:02Z)
Investigating the Robustness of Retrieval-Augmented Generation at the Query Level [4.3028340012580975]
Retrieval-augmented generation (RAG) has been proposed as a solution that dynamically incorporates external knowledge during inference.<n>Despite its promise, RAG systems face practical challenges-most notably, a strong dependence on the quality of the input query for accurate retrieval.
arXiv Detail & Related papers (2025-07-09T15:39:17Z)
RAG-Zeval: Towards Robust and Interpretable Evaluation on RAG Responses through End-to-End Rule-Guided Reasoning [64.46921169261852]
RAG-Zeval is a novel end-to-end framework that formulates faithfulness and correctness evaluation as a rule-guided reasoning task.<n>Our approach trains evaluators with reinforcement learning, facilitating compact models to generate comprehensive and sound assessments.<n>Experiments demonstrate RAG-Zeval's superior performance, achieving the strongest correlation with human judgments.
arXiv Detail & Related papers (2025-05-28T14:55:33Z)
Improving Multilingual Retrieval-Augmented Language Models through Dialectic Reasoning Argumentations [65.11348389219887]
We introduce Dialectic-RAG (DRAG), a modular approach that evaluates retrieved information by comparing, contrasting, and resolving conflicting perspectives.<n>We show the impact of our framework both as an in-context learning strategy and for constructing demonstrations to instruct smaller models.
arXiv Detail & Related papers (2025-04-07T06:55:15Z)
Unanswerability Evaluation for Retrieval Augmented Generation [74.3022365715597]
UAEval4RAG is a framework designed to evaluate whether RAG systems can handle unanswerable queries effectively.<n>We define a taxonomy with six unanswerable categories, and UAEval4RAG automatically synthesizes diverse and challenging queries.
arXiv Detail & Related papers (2024-12-16T19:11:55Z)
RAGChecker: A Fine-grained Framework for Diagnosing Retrieval-Augmented Generation [61.14660526363607]
We propose a fine-grained evaluation framework, RAGChecker, that incorporates a suite of diagnostic metrics for both the retrieval and generation modules. RAGChecker has significantly better correlations with human judgments than other evaluation metrics. The metrics of RAGChecker can guide researchers and practitioners in developing more effective RAG systems.
arXiv Detail & Related papers (2024-08-15T10:20:54Z)
RAGEval: Scenario Specific RAG Evaluation Dataset Generation Framework [66.93260816493553]
This paper introduces RAGEval, a framework designed to assess RAG systems across diverse scenarios.<n>With a focus on factual accuracy, we propose three novel metrics: Completeness, Hallucination, and Irrelevance.<n> Experimental results show that RAGEval outperforms zero-shot and one-shot methods in terms of clarity, safety, conformity, and richness of generated samples.
arXiv Detail & Related papers (2024-08-02T13:35:11Z)
Rethinking the Evaluation for Conversational Recommendation in the Era of Large Language Models [115.7508325840751]
The recent success of large language models (LLMs) has shown great potential to develop more powerful conversational recommender systems (CRSs) In this paper, we embark on an investigation into the utilization of ChatGPT for conversational recommendation, revealing the inadequacy of the existing evaluation protocol. We propose an interactive Evaluation approach based on LLMs named iEvaLM that harnesses LLM-based user simulators.
arXiv Detail & Related papers (2023-05-22T15:12:43Z)

This list is automatically generated from the titles and abstracts of the papers in this site.