Compare without Despair: Reliable Preference Evaluation with Generation Separability

• URL: http://arxiv.org/abs/2407.01878v3
• Date: Tue, 29 Oct 2024 17:29:37 GMT
• Title: Compare without Despair: Reliable Preference Evaluation with Generation Separability
• Authors: Sayan Ghosh, Tejas Srinivasan, Swabha Swayamdipta,
• Abstract summary: We introduce a measure, separability, which estimates how suitable a test instance is for pairwise preference evaluation. For a candidate test instance, separability samples multiple generations from a pair of models, and measures how distinguishable the two sets of generations are. Experiments show that instances with high separability values yield more consistent preference ratings from both human- and auto-raters.
• Score: 20.50638483427141
• License:
• Abstract: Human evaluation of generated language through pairwise preference judgments is pervasive. However, under common scenarios, such as when generations from a model pair are very similar, or when stochastic decoding results in large variations in generations, it results in inconsistent preference ratings. We address these challenges by introducing a meta-evaluation measure, separability, which estimates how suitable a test instance is for pairwise preference evaluation. For a candidate test instance, separability samples multiple generations from a pair of models, and measures how distinguishable the two sets of generations are. Our experiments show that instances with high separability values yield more consistent preference ratings from both human- and auto-raters. Further, the distribution of separability allows insights into which test benchmarks are more valuable for comparing models. Finally, we incorporate separability into ELO ratings, accounting for how suitable each test instance might be for reliably ranking LLMs. Overall, separability has implications for consistent, efficient and robust preference evaluation of LLMs with both human- and auto-raters.

Related papers

• The Comparative Trap: Pairwise Comparisons Amplifies Biased Preferences of LLM Evaluators [31.520403357740317]
  Large language models (LLMs) are increasingly used as evaluators for natural language generation tasks. LLMs display biased preferences, such as favoring verbosity and authoritative tones. We introduce PRePair, which integrates pointwise reasoning within a pairwise framework.
  arXiv  Detail & Related papers  (2024-06-18T06:43:04Z)
• Efficient LLM Comparative Assessment: a Product of Experts Framework for Pairwise Comparisons [10.94304714004328]
  This paper introduces a Product of Expert (PoE) framework for efficient Comparative Assessment. Individual comparisons are considered experts that provide information on a pair's score difference. PoE framework combines the information from these experts to yield an expression that can be maximized with respect to the underlying set of candidates.
  arXiv  Detail & Related papers  (2024-05-09T16:45:27Z)
• Examining the robustness of LLM evaluation to the distributional assumptions of benchmarks [2.1899189033259305]
  The research community often relies on a model's average performance across the test prompts of a benchmark to evaluate the model's performance. This is consistent with the assumption that the test prompts within a benchmark represent a random sample from a real-world distribution of interest. We find that (1) the correlation in model performance across test prompts is non-random, (2) accounting for correlations across test prompts can change model rankings on major benchmarks, and (3) explanatory factors for these correlations include semantic similarity and common LLM failure points.
  arXiv  Detail & Related papers  (2024-04-25T18:35:54Z)
• FairPair: A Robust Evaluation of Biases in Language Models through Paired Perturbations [33.24762796282484]
  We present FairPair, an evaluation framework for assessing differential treatment that occurs during ordinary usage. Unlike prior work, our method factors in the inherent variability that comes from the generation process itself by measuring the sampling variability.
  arXiv  Detail & Related papers  (2024-04-09T21:09:22Z)
• Aligning with Human Judgement: The Role of Pairwise Preference in Large Language Model Evaluators [48.54465599914978]
  Large Language Models (LLMs) have demonstrated promising capabilities in assessing the quality of generated natural language. LLMs still exhibit biases in evaluation and often struggle to generate coherent evaluations that align with human assessments. We introduce Pairwise-preference Search (PairS), an uncertainty-guided search method that employs LLMs to conduct pairwise comparisons and efficiently ranks candidate texts.
  arXiv  Detail & Related papers  (2024-03-25T17:11:28Z)
• A structured regression approach for evaluating model performance across intersectional subgroups [53.91682617836498]
  Disaggregated evaluation is a central task in AI fairness assessment, where the goal is to measure an AI system's performance across different subgroups. We introduce a structured regression approach to disaggregated evaluation that we demonstrate can yield reliable system performance estimates even for very small subgroups.
  arXiv  Detail & Related papers  (2024-01-26T14:21:45Z)
• 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)
• GREAT Score: Global Robustness Evaluation of Adversarial Perturbation using Generative Models [60.48306899271866]
  We present a new framework, called GREAT Score, for global robustness evaluation of adversarial perturbation using generative models. We show high correlation and significantly reduced cost of GREAT Score when compared to the attack-based model ranking on RobustBench. GREAT Score can be used for remote auditing of privacy-sensitive black-box models.
  arXiv  Detail & Related papers  (2023-04-19T14:58:27Z)
• 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)
• Two-Sample Testing on Ranked Preference Data and the Role of Modeling Assumptions [57.77347280992548]
  In this paper, we design two-sample tests for pairwise comparison data and ranking data. Our test requires essentially no assumptions on the distributions. By applying our two-sample test on real-world pairwise comparison data, we conclude that ratings and rankings provided by people are indeed distributed differently.
  arXiv  Detail & Related papers  (2020-06-21T20:51:09Z)

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.