Mitigating Label Bias with Interpretable Rubric Embeddings

• URL: http://arxiv.org/abs/2605.21455v1
• Date: Wed, 20 May 2026 17:46:08 GMT
• Title: Mitigating Label Bias with Interpretable Rubric Embeddings
• Abstract summary: We propose a representation framework that replaces standard black-box embeddings with features derived from expert-defined criteria.<n>We find that models trained on rubric embeddings reduce group disparities while improving measures of cohort quality.<n>Our results suggest that basing predictions on interpretable, domain-grounded representations offers a practical approach to learning in the presence of biased labels.
• Score: 2.58435220827336
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Statistical decision algorithms are increasingly deployed in domains where ground-truth labels are hard to obtain, such as hiring, university admissions, and content moderation. In these settings, models are typically trained on historical human evaluations -- for example, using past hiring decisions as a proxy for true applicant quality. However, if past evaluations unjustly favor certain groups, models trained on these labels may inherit those biases. To address this problem, we propose basing predictions on rubric embeddings, a representation framework that replaces standard black-box embeddings with features derived from expert-defined criteria that align with the underlying construct of interest. By anchoring predictions to semantically meaningful dimensions, this approach guards against biased proxy signals. We provide both theoretical and empirical evidence that rubric embeddings mitigate label bias under plausible conditions. Empirically, we evaluate our method on a novel dataset of applications to a large master's program. We find that models trained on rubric embeddings reduce group disparities while improving measures of cohort quality. Our results suggest that basing predictions on interpretable, domain-grounded representations offers a practical approach to learning in the presence of biased labels.

Related papers

• Bias Leaves a Gradient Trail: Label-Free Bias Identification via Gradient Probes on Concept Decompositions [8.894428198928333]
  We present a bias-label-free, post-hoc method for identifying spurious concepts in frozen vision models.<n>On Colored MNIST and Waterbirds the method recovers concepts aligned with the known spurious cue.<n> suppressing the top-ranked concepts at inference time improves worst-group accuracy by up to 17.9 percentage points.
  arXiv  Detail & Related papers  (2026-05-27T17:39:10Z)
• Unlabeled Debiasing in Downstream Tasks via Class-wise Low Variance Regularization [13.773597081543185]
  We introduce a novel debiasing regularization technique based on the class-wise variance of embeddings. Our method does not require attribute labels and targets any attribute, thus addressing the shortcomings of existing debiasing methods.
  arXiv  Detail & Related papers  (2024-09-29T03:56:50Z)
• Thinking Racial Bias in Fair Forgery Detection: Models, Datasets and Evaluations [63.52709761339949]
  We first contribute a dedicated dataset called the Fair Forgery Detection (FairFD) dataset, where we prove the racial bias of public state-of-the-art (SOTA) methods.<n>We design novel metrics including Approach Averaged Metric and Utility Regularized Metric, which can avoid deceptive results.<n>We also present an effective and robust post-processing technique, Bias Pruning with Fair Activations (BPFA), which improves fairness without requiring retraining or weight updates.
  arXiv  Detail & Related papers  (2024-07-19T14:53:18Z)
• Language-guided Detection and Mitigation of Unknown Dataset Bias [23.299264313976213]
  We propose a framework to identify potential biases as keywords without prior knowledge based on the partial occurrence in the captions. Our framework not only outperforms existing methods without prior knowledge, but also is even comparable with a method that assumes prior knowledge.
  arXiv  Detail & Related papers  (2024-06-05T03:11:33Z)
• Learning with Complementary Labels Revisited: The Selected-Completely-at-Random Setting Is More Practical [66.57396042747706]
  Complementary-label learning is a weakly supervised learning problem. We propose a consistent approach that does not rely on the uniform distribution assumption. We find that complementary-label learning can be expressed as a set of negative-unlabeled binary classification problems.
  arXiv  Detail & Related papers  (2023-11-27T02:59:17Z)
• Capturing Perspectives of Crowdsourced Annotators in Subjective Learning Tasks [9.110872603799839]
  Supervised classification heavily depends on datasets annotated by humans. In subjective tasks such as toxicity classification, these annotations often exhibit low agreement among raters. In this work, we propose textbfAnnotator Awares for Texts (AART) for subjective classification tasks.
  arXiv  Detail & Related papers  (2023-11-16T10:18:32Z)
• Evaluating the Fairness of Discriminative Foundation Models in Computer Vision [51.176061115977774]
  We propose a novel taxonomy for bias evaluation of discriminative foundation models, such as Contrastive Language-Pretraining (CLIP) We then systematically evaluate existing methods for mitigating bias in these models with respect to our taxonomy. Specifically, we evaluate OpenAI's CLIP and OpenCLIP models for key applications, such as zero-shot classification, image retrieval and image captioning.
  arXiv  Detail & Related papers  (2023-10-18T10:32:39Z)
• 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)
• On Comparing Fair Classifiers under Data Bias [42.43344286660331]
  We study the effect of varying data biases on the accuracy and fairness of fair classifiers. Our experiments show how to integrate a measure of data bias risk in the existing fairness dashboards for real-world deployments.
  arXiv  Detail & Related papers  (2023-02-12T13:04:46Z)
• Systematic Evaluation of Predictive Fairness [60.0947291284978]
  Mitigating bias in training on biased datasets is an important open problem. We examine the performance of various debiasing methods across multiple tasks. We find that data conditions have a strong influence on relative model performance.
  arXiv  Detail & Related papers  (2022-10-17T05:40:13Z)
• Social Bias Meets Data Bias: The Impacts of Labeling and Measurement Errors on Fairness Criteria [4.048444203617942]
  We consider two forms of dataset bias: errors by prior decision makers in the labeling process, and errors in measurement of the features of disadvantaged individuals. We analytically show that some constraints can remain robust when facing certain statistical biases, while others (such as Equalized Odds) are significantly violated if trained on biased data. Our findings present an additional guideline for choosing among existing fairness criteria, or for proposing new criteria, when available datasets may be biased.
  arXiv  Detail & Related papers  (2022-05-31T22:43:09Z)
• Optimising Equal Opportunity Fairness in Model Training [60.0947291284978]
  Existing debiasing methods, such as adversarial training and removing protected information from representations, have been shown to reduce bias. We propose two novel training objectives which directly optimise for the widely-used criterion of it equal opportunity, and show that they are effective in reducing bias while maintaining high performance over two classification tasks.
  arXiv  Detail & Related papers  (2022-05-05T01:57:58Z)
• Balancing out Bias: Achieving Fairness Through Training Reweighting [58.201275105195485]
  Bias in natural language processing arises from models learning characteristics of the author such as gender and race. Existing methods for mitigating and measuring bias do not directly account for correlations between author demographics and linguistic variables. This paper introduces a very simple but highly effective method for countering bias using instance reweighting.
  arXiv  Detail & Related papers  (2021-09-16T23:40:28Z)

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.