GroUSE: A Benchmark to Evaluate Evaluators in Grounded Question Answering

• URL: http://arxiv.org/abs/2409.06595v1
• Date: Tue, 10 Sep 2024 15:39:32 GMT
• Title: GroUSE: A Benchmark to Evaluate Evaluators in Grounded Question Answering
• Authors: Sacha Muller, António Loison, Bilel Omrani, Gautier Viaud,
• Abstract summary: Retrieval-Augmented Generation (RAG) has emerged as a common paradigm to use Large Language Models (LLMs) alongside private and up-to-date knowledge bases. We address the challenges of using LLM-as-a-Judge when evaluating grounded answers generated by RAG systems.
• Score: 0.0
• License: http://creativecommons.org/publicdomain/zero/1.0/
• Abstract: Retrieval-Augmented Generation (RAG) has emerged as a common paradigm to use Large Language Models (LLMs) alongside private and up-to-date knowledge bases. In this work, we address the challenges of using LLM-as-a-Judge when evaluating grounded answers generated by RAG systems. To assess the calibration and discrimination capabilities of judge models, we identify 7 generator failure modes and introduce GroUSE (Grounded QA Unitary Scoring of Evaluators), a meta-evaluation benchmark of 144 unit tests. This benchmark reveals that existing automated RAG evaluation frameworks often overlook important failure modes, even when using GPT-4 as a judge. To improve on the current design of automated RAG evaluation frameworks, we propose a novel pipeline and find that while closed models perform well on GroUSE, state-of-the-art open-source judges do not generalize to our proposed criteria, despite strong correlation with GPT-4's judgement. Our findings suggest that correlation with GPT-4 is an incomplete proxy for the practical performance of judge models and should be supplemented with evaluations on unit tests for precise failure mode detection. We further show that finetuning Llama-3 on GPT-4's reasoning traces significantly boosts its evaluation capabilities, improving upon both correlation with GPT-4's evaluations and calibration on reference situations.

Related papers

• Language Model Preference Evaluation with Multiple Weak Evaluators [78.53743237977677]
  GED (Preference Graph Ensemble and Denoise) is a novel approach that leverages multiple model-based evaluators to construct preference graphs. We show that GED outperforms baseline methods in model ranking, response selection, and model alignment tasks.
  arXiv  Detail & Related papers  (2024-10-14T01:57:25Z)
• 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)
• RankRAG: Unifying Context Ranking with Retrieval-Augmented Generation in LLMs [60.38044044203333]
  Large language models (LLMs) typically utilize the top-k contexts from a retriever in retrieval-augmented generation (RAG) We propose a novel instruction fine-tuning framework RankRAG, which instruction-tunes a single LLM for the dual purpose of context ranking and answer generation in RAG. For generation, we compare our model with many strong baselines, including GPT-4-0613, GPT-4-turbo-2024-0409, and ChatQA-1.5, an open-sourced model with the state-of-the-art performance on RAG benchmarks.
  arXiv  Detail & Related papers  (2024-07-02T17:59:17Z)
• An Empirical Study of LLM-as-a-Judge for LLM Evaluation: Fine-tuned Judge Model is not a General Substitute for GPT-4 [29.93673872618022]
  Fine-tuned judge models are claimed to achieve comparable evaluation capability with GPT-4. We introduce a method, leveraging GPT-4 to compensate for the limitations and improve the fine-tuned judges. Experiment results show our method achieves accuracy on par with GPT-4 with only 50% of the API expense.
  arXiv  Detail & Related papers  (2024-03-05T10:20:52Z)
• ChatQA: Surpassing GPT-4 on Conversational QA and RAG [43.34692996785167]
  We introduce ChatQA, a suite of models that outperform GPT-4 on retrieval-augmented generation (RAG) and conversational question answering (QA) For effective retrieval, we introduce a dense retriever optimized for conversational QA, which yields results comparable to the alternative state-of-the-art query rewriting models. We present the ChatRAG Bench, which encompasses ten datasets covering comprehensive evaluations on RAG, table-related QA, arithmetic calculations, and scenarios involving unanswerable questions.
  arXiv  Detail & Related papers  (2024-01-18T18:59:11Z)
• CritiqueLLM: Towards an Informative Critique Generation Model for Evaluation of Large Language Model Generation [87.44350003888646]
  Eval-Instruct can acquire pointwise grading critiques with pseudo references and revise these critiques via multi-path prompting. CritiqueLLM is empirically shown to outperform ChatGPT and all the open-source baselines.
  arXiv  Detail & Related papers  (2023-11-30T16:52:42Z)
• Evaluation Metrics in the Era of GPT-4: Reliably Evaluating Large Language Models on Sequence to Sequence Tasks [9.801767683867125]
  We provide a preliminary and hybrid evaluation on three NLP benchmarks using both automatic and human evaluation. We find that ChatGPT consistently outperforms many other popular models according to human reviewers on the majority of metrics. We also find that human reviewers rate the gold reference as much worse than the best models' outputs, indicating the poor quality of many popular benchmarks.
  arXiv  Detail & Related papers  (2023-10-20T20:17:09Z)
• Is GPT-4 a reliable rater? Evaluating Consistency in GPT-4 Text Ratings [63.35165397320137]
  This study investigates the consistency of feedback ratings generated by OpenAI's GPT-4. The model rated responses to tasks within the Higher Education subject domain of macroeconomics in terms of their content and style.
  arXiv  Detail & Related papers  (2023-08-03T12:47:17Z)
• ARB: Advanced Reasoning Benchmark for Large Language Models [94.37521840642141]
  We introduce ARB, a novel benchmark composed of advanced reasoning problems in multiple fields. As a subset of ARB, we introduce a challenging set of math and physics problems which require advanced symbolic reasoning and domain knowledge. We evaluate recent models such as GPT-4 and Claude on ARB and demonstrate that current models score well below 50% on more demanding tasks.
  arXiv  Detail & Related papers  (2023-07-25T17:55:19Z)
• Analyzing the Performance of GPT-3.5 and GPT-4 in Grammatical Error Correction [28.58384091374763]
  GPT-3 and GPT-4 models are powerful, achieving high performance on a variety of Natural Language Processing tasks. We perform experiments testing the capabilities of a GPT-3.5 model (text-davinci-003) and a GPT-4 model (gpt-4-0314) on major GEC benchmarks. We report the performance of our best prompt on the BEA-2019 and JFLEG datasets, finding that the GPT models can perform well in a sentence-level revision setting.
  arXiv  Detail & Related papers  (2023-03-25T03:08:49Z)

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

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.