Beyond correlation: The Impact of Human Uncertainty in Measuring the Effectiveness of Automatic Evaluation and LLM-as-a-Judge

• URL: http://arxiv.org/abs/2410.03775v3
• Date: Mon, 27 Jan 2025 07:02:04 GMT
• Title: Beyond correlation: The Impact of Human Uncertainty in Measuring the Effectiveness of Automatic Evaluation and LLM-as-a-Judge
• Authors: Aparna Elangovan, Lei Xu, Jongwoo Ko, Mahsa Elyasi, Ling Liu, Sravan Bodapati, Dan Roth,
• Abstract summary: We show how *relying on a single aggregate correlation score* can obscure fundamental differences between human labels and those from automatic evaluation. We propose stratifying data by human label uncertainty to provide a more robust analysis of automatic evaluation performance.
• Score: 51.93909886542317
• License:
• Abstract: The effectiveness of automatic evaluation of generative models is typically measured by comparing the labels generated via automation with labels by humans using correlation metrics. However, metrics like Krippendorff's $\alpha$ and Randolph's $\kappa$ were originally designed to measure the reliability of human labeling, thus make assumptions about typical human labeling behavior, and these assumptions may not be applicable to machine generated labels. In this paper, we show how *relying on a single aggregate correlation score* can obscure fundamental differences between human labels and those from automatic evaluation, including LLM-as-a-Judge. Specifically, we demonstrate that when the proportion of samples with variation or uncertainty in human assigned labels is relatively high, machine labels (generated by automatic evaluation methods) may superficially appear to have similar or better correlation with the human majority label compared to the human-to-human (HH) correlation. This can create the illusion that labels from automatic evaluation approximates the human majority label. However, as the proportion of samples with consistent human labels increases, the correlation between machine and human labels fall well below HH correlation. Based on these findings, we first propose stratifying data by human label uncertainty to provide a more robust analysis of automatic evaluation performance. Second, recognizing that uncertainty and variation are inherent in perception-based human evaluations, such as those involving attitudes or preferences, we introduce a new metric - binned Jensen-Shannon Divergence for perception for such scenarios to better measure the effectiveness of automatic evaluations. We present visualization techniques -- perception charts, to contextualize correlation measures appropriately. We have open-sourced at https://github.com/amazon-science/BeyondCorrelation.

Related papers

• "All that Glitters": Approaches to Evaluations with Unreliable Model and Human Annotations [0.0]
  "Gold" and "ground truth" human-mediated labels have error. This study demonstrates methods for answering such questions even in the context of very low reliabilities from expert humans.
  arXiv  Detail & Related papers  (2024-11-23T19:18:08Z)
• Learning with Confidence: Training Better Classifiers from Soft Labels [0.0]
  In supervised machine learning, models are typically trained using data with hard labels, i.e., definite assignments of class membership. We investigate whether incorporating label uncertainty, represented as discrete probability distributions over the class labels, improves the predictive performance of classification models.
  arXiv  Detail & Related papers  (2024-09-24T13:12:29Z)
• 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)
• Gender Biases in Automatic Evaluation Metrics for Image Captioning [87.15170977240643]
  We conduct a systematic study of gender biases in model-based evaluation metrics for image captioning tasks. We demonstrate the negative consequences of using these biased metrics, including the inability to differentiate between biased and unbiased generations. We present a simple and effective way to mitigate the metric bias without hurting the correlations with human judgments.
  arXiv  Detail & Related papers  (2023-05-24T04:27:40Z)
• Improving Classifier Robustness through Active Generation of Pairwise Counterfactuals [22.916599410472102]
  We present a novel framework that utilizes counterfactual generative models to generate a large number of diverse counterfactuals. We show that with a small amount of human-annotated counterfactual data (10%), we can generate a counterfactual augmentation dataset with learned labels.
  arXiv  Detail & Related papers  (2023-05-22T23:19:01Z)
• The 'Problem' of Human Label Variation: On Ground Truth in Data, Modeling and Evaluation [21.513743126525622]
  We argue that this big open problem of human label variation persists and critically needs more attention to move our field forward. We reconcile different previously proposed notions of human label variation, provide a repository of publicly-available datasets with un-aggregated labels, depict approaches proposed so far, identify gaps and suggest ways forward.
  arXiv  Detail & Related papers  (2022-11-04T16:38:09Z)
• Multi-label Classification with High-rank and High-order Label Correlations [62.39748565407201]
  Previous methods capture the high-order label correlations mainly by transforming the label matrix to a latent label space with low-rank matrix factorization. We propose a simple yet effective method to depict the high-order label correlations explicitly, and at the same time maintain the high-rank of the label matrix. Comparative studies over twelve benchmark data sets validate the effectiveness of the proposed algorithm in multi-label classification.
  arXiv  Detail & Related papers  (2022-07-09T05:15:31Z)
• Re-Examining System-Level Correlations of Automatic Summarization Evaluation Metrics [64.81682222169113]
  How reliably an automatic summarization evaluation metric replicates human judgments of summary quality is quantified by system-level correlations. We identify two ways in which the definition of the system-level correlation is inconsistent with how metrics are used to evaluate systems in practice.
  arXiv  Detail & Related papers  (2022-04-21T15:52:14Z)
• An Empirical Investigation of Learning from Biased Toxicity Labels [15.822714574671412]
  We study how different training strategies can leverage a small dataset of human-annotated labels and a large but noisy dataset of synthetically generated labels. We evaluate the accuracy and fairness properties of these approaches, and trade-offs between the two.
  arXiv  Detail & Related papers  (2021-10-04T17:19:57Z)
• Exploiting Sample Uncertainty for Domain Adaptive Person Re-Identification [137.9939571408506]
  We estimate and exploit the credibility of the assigned pseudo-label of each sample to alleviate the influence of noisy labels. Our uncertainty-guided optimization brings significant improvement and achieves the state-of-the-art performance on benchmark datasets.
  arXiv  Detail & Related papers  (2020-12-16T04:09:04Z)

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.