Improving Reproducibility in Evaluation through Multi-Level Annotator Modeling

• URL: http://arxiv.org/abs/2605.13801v1
• Date: Wed, 13 May 2026 17:22:27 GMT
• Title: Improving Reproducibility in Evaluation through Multi-Level Annotator Modeling
• Abstract summary: We introduce a multi-level bootstrapping approach to realistically model annotator behavior.<n>We analyze the tradeoffs between the number of items ($N$) and the number of responses per item ($K$) required to achieve statistical significance.
• Score: 6.372025352636228
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: As generative AI models such as large language models (LLMs) become more pervasive, ensuring the safety, robustness, and overall trustworthiness of these systems is paramount. However, AI is currently facing a reproducibility crisis driven by unreliable evaluations and unrepeatable experimental results. While human raters are often used to assess models for utility and safety, they introduce divergent biases and subjective opinions into their annotations. Overcoming this variance is exceptionally challenging because very little data exists to study how experimental repeatability actually improves as the annotator pool grows. Standard evaluation practices typically rely on a small number of annotations per item (often 3 to 5) and lack the persistent rater identifiers necessary to model individual variance across items. In this work, we introduce a multi-level bootstrapping approach to realistically model annotator behavior. Leveraging datasets with a large number of ratings and persistent rater identifiers, we analyze the tradeoffs between the number of items ($N$) and the number of responses per item ($K$) required to achieve statistical significance.

Related papers

• Debiasing Reward Models by Representation Learning with Guarantees [20.170532007501883]
  We propose a principled framework that mitigates spurious correlation issues in reward models.<n>We show that non-spurious latent variables can be theoretically identified from data, regardless of whether a surrogate for the spurious latent variables is available.<n>Experiments on synthetic and real-world datasets demonstrate that our method effectively mitigates spurious correlation issues and yields more robust reward models.
  arXiv  Detail & Related papers  (2025-10-27T18:37:57Z)
• Adaptive Generation of Bias-Eliciting Questions for LLMs [18.608477560948003]
  Large language models (LLMs) are now widely deployed in user-facing applications, reaching hundreds of millions worldwide.<n>We introduce a counterfactual bias evaluation framework that automatically generates realistic, open-ended questions over sensitive attributes such as sex, race, or religion.<n>We also capture distinct response dimensions that are increasingly relevant in user interactions, such as asymmetric refusals and explicit acknowledgment of bias.
  arXiv  Detail & Related papers  (2025-10-14T13:08:10Z)
• Self-Consistency as a Free Lunch: Reducing Hallucinations in Vision-Language Models via Self-Reflection [71.8243083897721]
  Vision-language models often hallucinate details, generating non-existent objects or inaccurate attributes that compromise output reliability.<n>We present a novel framework that leverages the model's self-consistency between long responses and short answers to generate preference pairs for training.
  arXiv  Detail & Related papers  (2025-09-27T10:37:11Z)
• Preference Learning for AI Alignment: a Causal Perspective [55.2480439325792]
  We frame this problem in a causal paradigm, providing the rich toolbox of causality to identify persistent challenges.<n>Inheriting from the literature of causal inference, we identify key assumptions necessary for reliable generalisation.<n>We illustrate failure modes of naive reward models and demonstrate how causally-inspired approaches can improve model robustness.
  arXiv  Detail & Related papers  (2025-06-06T10:45:42Z)
• How Many Ratings per Item are Necessary for Reliable Significance Testing? [7.422152765037947]
  A cornerstone of machine learning evaluation is the assumption that model and human responses are reliable enough to evaluate models against unitary, authoritative, gold standard'' data.<n>We adapt a method to determine whether an (existing or planned) dataset has enough responses per item to assure reliable null hypothesis statistical testing.<n>We show how our methods can help AI researchers make better decisions about how to collect data for AI evaluation.
  arXiv  Detail & Related papers  (2024-12-04T02:31:28Z)
• Regression for the Mean: Auto-Evaluation and Inference with Few Labels through Post-hoc Regression [4.813376208491175]
  The Prediction Powered Inference (PPI) framework provides a way of leveraging both a large pool of pseudo-labelled data and a small sample with real, high-quality labels.<n>We find that when labelled data is scarce, the PPI++ method can perform even worse than classical inference.<n>We present two new PPI-based techniques that leverage robust regressors to produce even lower variance estimators in the few-label regime.
  arXiv  Detail & Related papers  (2024-11-19T17:17:46Z)
• 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)
• 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)
• Rethinking Missing Data: Aleatoric Uncertainty-Aware Recommendation [59.500347564280204]
  We propose a new Aleatoric Uncertainty-aware Recommendation (AUR) framework. AUR consists of a new uncertainty estimator along with a normal recommender model. As the chance of mislabeling reflects the potential of a pair, AUR makes recommendations according to the uncertainty.
  arXiv  Detail & Related papers  (2022-09-22T04:32:51Z)
• CausalAgents: A Robustness Benchmark for Motion Forecasting using Causal Relationships [8.679073301435265]
  We construct a new benchmark for evaluating and improving model robustness by applying perturbations to existing data. We use these labels to perturb the data by deleting non-causal agents from the scene. Under non-causal perturbations, we observe a $25$-$38%$ relative change in minADE as compared to the original.
  arXiv  Detail & Related papers  (2022-07-07T21:28:23Z)
• Accurate and Robust Feature Importance Estimation under Distribution Shifts [49.58991359544005]
  PRoFILE is a novel feature importance estimation method. We show significant improvements over state-of-the-art approaches, both in terms of fidelity and robustness.
  arXiv  Detail & Related papers  (2020-09-30T05:29:01Z)

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.