STABLEVAL: Disagreement-Aware and Stable Evaluation of AI Systems

• URL: http://arxiv.org/abs/2605.02122v1
• Date: Mon, 04 May 2026 01:03:48 GMT
• Title: STABLEVAL: Disagreement-Aware and Stable Evaluation of AI Systems
• Abstract summary: Majority vote discards annotator reliability and item-level ambiguity, often yielding unstable comparisons.<n>We introduce STABLEVAL, a disagreement-aware evaluation framework.<n>STABLEVAL is explicitly designed for stable and uncertainty-aware system evaluation.
• Score: 3.263813419432528
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Human evaluation remains the primary standard for assessing modern AI systems, yet annotator disagreement, bias, and variability make system rankings fragile under standard majority vote aggregation. Majority vote discards annotator reliability and item-level ambiguity, often yielding unstable comparisons across annotator subsets. We introduce STABLEVAL, a disagreement-aware evaluation framework that models latent item correctness and annotator-specific confusion patterns to produce posterior expected item credit and calibrated agent-level scores. Unlike label-denoising approaches such as Dawid-Skene, STABLEVAL is explicitly designed for stable and uncertainty-aware system evaluation rather than hard label recovery. We formalize ranking stability as a first-class evaluation objective and analyze how aggregation methods preserve or distort underlying annotator behavior. Across controlled synthetic experiments and multiple real-world human-annotated benchmarks, majority vote exhibits increasing score error and ranking instability under annotator heterogeneity and adversarial noise, while STABLEVAL yields more stable and statistically grounded system rankings. These results demonstrate that modeling disagreement is essential for robust and reproducible AI evaluation.

Related papers

• Position: Evaluation Scores Are Perishable Knowledge Claims [0.0]
  We argue that evaluation scores should be treated as claims with three properties: formality, scope, and validity windows.<n>We illustrate the cost of mean aggregation on the public HELM leaderboard.
  arXiv  Detail & Related papers  (2026-07-28T18:50:39Z)
• On the Stability of Prompt Ranking in Large Language Model Evaluation [9.718200364661675]
  This paper systematically studies prompt ranking stability under common sources of variability, including random seeds and limited evaluation subsets.<n>We find that while overall rank correlations are often moderate to high, the identity of the top-performing prompt frequently changes, leading to unreliable selection decisions.<n>We propose a simple stability-aware selection strategy based on a lower confidence bound, which accounts for both performance and variance.
  arXiv  Detail & Related papers  (2026-06-23T10:13:47Z)
• A Unified Perturbation Framework for Analyzing Leaderboard Stability and Manipulation [47.21240079171094]
  We present a unified perturbation framework for analyzing Bradley-Terry leaderboards.<n>Our framework studies three match-level perturbations -- Drop, Add, and Flip.<n>We show that modern leaderboards are non-robust across all three objectives.
  arXiv  Detail & Related papers  (2026-05-15T09:21:33Z)
• Robust Checkpoint Selection for Multimodal LLMs via Agentic Evaluation and Stability-Aware Ranking [0.4562811934651716]
  We formulate checkpoint selection as a robust decision problem under evaluation uncertainty.<n>We propose a multi-stage framework that integrates curated real-world data, structured LLM-based judgment, and multi-stage ranking protocols.<n>We demonstrate that data quality, specifically OCR readability, is a critical determinant of evaluation validity.
  arXiv  Detail & Related papers  (2026-05-13T12:18:32Z)
• Multimodal Learning on Low-Quality Data with Conformal Predictive Self-Calibration [72.0672328514289]
  Multimodal learning often grapples with the challenge of low-quality data.<n>We propose a unified framework, termed Conformal Predictive Self-Calibration.<n>Our framework consistently outperforms existing state-of-the-art methods.
  arXiv  Detail & Related papers  (2026-05-05T14:48:52Z)
• Towards More Standardized AI Evaluation: From Models to Agents [0.0]
  As AI systems evolve from static models to compound, tool-using agents, evaluation becomes a core control function.<n>Most evaluation practices remain anchored in assumptions inherited from the model-centric era.<n>This paper argues that such approaches are increasingly obscure rather than illuminating system behavior.
  arXiv  Detail & Related papers  (2026-02-20T06:54:44Z)
• The Emergence of Lab-Driven Alignment Signatures: A Psychometric Framework for Auditing Latent Bias and Compounding Risk in Generative AI [0.0]
  This paper introduces a novel auditing framework to quantify latent trait estimation under ordinal uncertainty.<n>The research audits nine leading models across dimensions including Optimization Bias, Sycophancy, and Status-Quo Legitimization.
  arXiv  Detail & Related papers  (2026-02-19T06:56:01Z)
• Not All Preferences Are Created Equal: Stability-Aware and Gradient-Efficient Alignment for Reasoning Models [52.48582333951919]
  We propose a dynamic framework designed to enhance alignment reliability by maximizing the Signal-to-Noise Ratio of policy updates.<n>SAGE (Stability-Aware Gradient Efficiency) integrates a coarse-grained curriculum mechanism that refreshes candidate pools based on model competence.<n> Experiments on multiple mathematical reasoning benchmarks demonstrate that SAGE significantly accelerates convergence and outperforms static baselines.
  arXiv  Detail & Related papers  (2026-02-01T12:56:10Z)
• Reference-Free Rating of LLM Responses via Latent Information [53.463883683503106]
  We study the common practice of asking a judge model to assign Likert-scale scores to free-text responses.<n>We then propose and evaluate Latent Judges, which derive scalar ratings from internal model signals.<n>Across a broad suite of pairwise and single-rating benchmarks, latent methods match or surpass standard prompting.
  arXiv  Detail & Related papers  (2025-09-29T12:15:52Z)
• SVeritas: Benchmark for Robust Speaker Verification under Diverse Conditions [54.34001921326444]
  Speaker verification (SV) models are increasingly integrated into security, personalization, and access control systems.<n>Existing benchmarks evaluate only subsets of these conditions, missing others entirely.<n>We introduce SVeritas, a comprehensive Speaker Verification tasks benchmark suite, assessing SV systems under stressors like recording duration, spontaneity, content, noise, microphone distance, reverberation, channel mismatches, audio bandwidth, codecs, speaker age, and susceptibility to spoofing and adversarial attacks.
  arXiv  Detail & Related papers  (2025-09-21T14:11:16Z)
• UDA: Unsupervised Debiasing Alignment for Pair-wise LLM-as-a-Judge [23.497453639857852]
  We propose UDA, a framework that reduces inter-judge disagreement by dynamically adjusting the Elo rating system.<n>UDA operates in a fully unsupervised manner, guided solely by the objective of minimizing the dispersion among the Elo trajectories of all judges.<n>Experiments show that UDA significantly reduces the inter-judge rating standard deviation by up to 63.4% and improves the average correlation with human judgments by 24.7%.
  arXiv  Detail & Related papers  (2025-08-13T11:41:01Z)
• Position: AI Evaluation Should Learn from How We Test Humans [65.36614996495983]
  We argue that psychometrics, a theory originating in the 20th century for human assessment, could be a powerful solution to the challenges in today's AI evaluations.
  arXiv  Detail & Related papers  (2023-06-18T09:54:33Z)
• 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.