How Reliable Are Automatic Evaluation Methods for Instruction-Tuned LLMs?

• URL: http://arxiv.org/abs/2402.10770v4
• Date: Wed, 02 Oct 2024 09:18:47 GMT
• Title: How Reliable Are Automatic Evaluation Methods for Instruction-Tuned LLMs?
• Authors: Ehsan Doostmohammadi, Oskar Holmström, Marco Kuhlmann,
• Abstract summary: We perform a meta-evaluation of such methods and assess their reliability across a broad range of tasks. We observe that while automatic evaluation methods can approximate human ratings under specific conditions, their validity is highly context-dependent. Our findings enhance the understanding of how automatic methods should be applied and interpreted when developing and evaluating instruction-tuned LLMs.
• Score: 3.1706553206969925
• License:
• Abstract: Work on instruction-tuned Large Language Models (LLMs) has used automatic methods based on text overlap and LLM judgments as cost-effective alternatives to human evaluation. In this paper, we perform a meta-evaluation of such methods and assess their reliability across a broad range of tasks. In evaluating how well automatic methods align with human evaluations, correlation metrics are the most commonly employed method despite their inherent limitations when dealing with ties and different scales. To address these shortcomings, we use Pairwise Accuracy as an alternative to standard correlation measures. We observe that while automatic evaluation methods can approximate human ratings under specific conditions, their validity is highly context-dependent. Specifically, the simple ROUGE-L metric correlates very well with human ratings for short-answer English tasks but is unreliable in free-form generation tasks and cross-lingual scenarios. The effectiveness of the more advanced method of using GPT-4 as a judge diminishes significantly if reference answers are not included in the prompt, which is the scenario where this method has the potential to provide the most value compared to other metrics. Our findings enhance the understanding of how automatic methods should be applied and interpreted when developing and evaluating instruction-tuned LLMs.

Related papers

• Reference-Guided Verdict: LLMs-as-Judges in Automatic Evaluation of Free-Form Text [12.879551933541345]
  Large Language Models (LLMs) are capable of generating human-like conversations. Conventional metrics like BLEU and ROUGE are inadequate for capturing the subtle semantics and contextual richness of such generative outputs. We propose a reference-guided verdict method that automates the evaluation process by leveraging multiple LLMs-as-judges.
  arXiv  Detail & Related papers  (2024-08-17T16:01:45Z)
• Decoding Biases: Automated Methods and LLM Judges for Gender Bias Detection in Language Models [47.545382591646565]
  Large Language Models (LLMs) have excelled at language understanding and generating human-level text. LLMs are susceptible to adversarial attacks where malicious users prompt the model to generate undesirable text. In this work, we train models to automatically create adversarial prompts to elicit biased responses from target LLMs.
  arXiv  Detail & Related papers  (2024-08-07T17:11:34Z)
• SemScore: Automated Evaluation of Instruction-Tuned LLMs based on Semantic Textual Similarity [3.3162484539136416]
  We propose a simple but remarkably effective evaluation metric called SemScore. We compare model outputs to gold target responses using semantic textual similarity (STS) We find that our proposed SemScore metric outperforms all other, in many cases more complex, evaluation metrics in terms of correlation to human evaluation.
  arXiv  Detail & Related papers  (2024-01-30T14:52:50Z)
• Self-Evaluation Improves Selective Generation in Large Language Models [54.003992911447696]
  We reformulate open-ended generation tasks into token-level prediction tasks. We instruct an LLM to self-evaluate its answers. We benchmark a range of scoring methods based on self-evaluation.
  arXiv  Detail & Related papers  (2023-12-14T19:09:22Z)
• Towards Better Evaluation of Instruction-Following: A Case-Study in Summarization [9.686937153317809]
  We perform a meta-evaluation of a variety of metrics to quantify how accurately they measure the instruction-following abilities of large language models. Using riSum, we analyze the agreement between evaluation methods and human judgment.
  arXiv  Detail & Related papers  (2023-10-12T15:07:11Z)
• Bring Your Own Data! Self-Supervised Evaluation for Large Language Models [52.15056231665816]
  We propose a framework for self-supervised evaluation of Large Language Models (LLMs) We demonstrate self-supervised evaluation strategies for measuring closed-book knowledge, toxicity, and long-range context dependence. We find strong correlations between self-supervised and human-supervised evaluations.
  arXiv  Detail & Related papers  (2023-06-23T17:59:09Z)
• Large Language Models are Not Yet Human-Level Evaluators for Abstractive Summarization [66.08074487429477]
  We investigate the stability and reliability of large language models (LLMs) as automatic evaluators for abstractive summarization. We find that while ChatGPT and GPT-4 outperform the commonly used automatic metrics, they are not ready as human replacements.
  arXiv  Detail & Related papers  (2023-05-22T14:58:13Z)
• TRUE: Re-evaluating Factual Consistency Evaluation [29.888885917330327]
  We introduce TRUE: a comprehensive study of factual consistency metrics on a standardized collection of existing texts from diverse tasks. Our standardization enables an example-level meta-evaluation protocol that is more actionable and interpretable than previously reported correlations. Across diverse state-of-the-art metrics and 11 datasets we find that large-scale NLI and question generation-and-answering-based approaches achieve strong and complementary results.
  arXiv  Detail & Related papers  (2022-04-11T10:14:35Z)
• A Statistical Analysis of Summarization Evaluation Metrics using Resampling Methods [60.04142561088524]
  We find that the confidence intervals are rather wide, demonstrating high uncertainty in how reliable automatic metrics truly are. Although many metrics fail to show statistical improvements over ROUGE, two recent works, QAEval and BERTScore, do in some evaluation settings.
  arXiv  Detail & Related papers  (2021-03-31T18:28:14Z)
• PONE: A Novel Automatic Evaluation Metric for Open-Domain Generative Dialogue Systems [48.99561874529323]
  There are three kinds of automatic methods to evaluate the open-domain generative dialogue systems. Due to the lack of systematic comparison, it is not clear which kind of metrics are more effective. We propose a novel and feasible learning-based metric that can significantly improve the correlation with human judgments.
  arXiv  Detail & Related papers  (2020-04-06T04:36:33Z)

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.