Learning Who Disagrees: Demographic Importance Weighting for Modeling Annotator Distributions with DiADEM

• URL: http://arxiv.org/abs/2604.08425v1
• Date: Thu, 09 Apr 2026 16:29:17 GMT
• Title: Learning Who Disagrees: Demographic Importance Weighting for Modeling Annotator Distributions with DiADEM
• Abstract summary: We introduce DiADEM, a neural architecture that learns "how much each demographic axis matters" for predicting who will disagree.<n>Our results demonstrate that explicitly modeling who annotators are not just what they label is essential for NLP systems.
• Score: 7.219237599532391
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: When humans label subjective content, they disagree, and that disagreement is not noise. It reflects genuine differences in perspective shaped by annotators' social identities and lived experiences. Yet standard practice still flattens these judgments into a single majority label, and recent LLM-based approaches fare no better: we show that prompted large language models, even with chain-of-thought reasoning, fail to recover the structure of human disagreement. We introduce DiADEM, a neural architecture that learns "how much each demographic axis matters" for predicting who will disagree and on what. DiADEM encodes annotators through per-demographic projections governed by a learned importance vector $\boldsymbolα$, fuses annotator and item representations via complementary concatenation and Hadamard interactions, and is trained with a novel item-level disagreement loss that directly penalizes mispredicted annotation variance. On the DICES conversational-safety and VOICED political-offense benchmarks, DiADEM substantially outperforms both the LLM-as-a-judge and neural model baselines across standard and perspectivist metrics, achieving strong disagreement tracking ($r{=}0.75$ on DICES). The learned $\boldsymbolα$ weights reveal that race and age consistently emerge as the most influential demographic factors driving annotator disagreement across both datasets. Our results demonstrate that explicitly modeling who annotators are not just what they label is essential for NLP systems that aim to faithfully represent human interpretive diversity.

Related papers

• Mitigating Bias in Large Vision-Language Models via Counterfactual Ensemble Decoding [45.1503984011004]
  Large Vision-Language Models (LVLMs) have achieved remarkable performance across a wide range of tasks.<n>They often inherit social biases from their training data, resulting in biased behavior when processing portraits from different social groups.<n>We propose Counterfactual Ensemble Decoding (CED), a novel framework that constructs multi-group counterfactual perspectives.
  arXiv  Detail & Related papers  (2026-08-12T08:25:29Z)
• Paying More Attention to Visual Tokens in Self-Evolving Large Multimodal Models [54.097381112138]
  Current self-evolving LMMs struggle on vision-language understanding tasks such as image captioning and visual question answering.<n>We propose VISE, a purely unsupervised self-evolving framework that directly regularizes the model's visual conditioning policy.<n>VISE operates within a single model without specialist roles, external reward models, or annotations.
  arXiv  Detail & Related papers  (2026-06-25T17:59:55Z)
• The Ghost Annotator: a Framework to Explore Human Label Variation in Content Moderation through Conformal Prediction [39.10305114526259]
  We introduce a framework combining conformal prediction with Collaborative Filtering-style annotators' representation.<n>Using Non-Conformity Scores, we introduce the Ghost Prediction metric and the Ghost Annotator representation to quantify cases in which model predictions diverge from all available human annotations.<n>Our finding shows that while all models uncertainty increases with annotator disagreement, larger models tend to be more confident in the classification of texts that are not aligned with any human annotation.
  arXiv  Detail & Related papers  (2026-06-01T21:32:37Z)
• Dependence-Aware Label Aggregation for LLM-as-a-Judge via Ising Models [55.94503936470247]
  Large-scale AI evaluation increasingly relies on aggregating binary judgments from $K$ annotators, including judges.<n>Most classical methods assume annotators are conditionally independent given the true label $Yin0,1$, an assumption often violated by LLM judges.<n>We study label aggregation through a hierarchy of dependence-aware models based on Ising graphical models and latent factors.
  arXiv  Detail & Related papers  (2026-01-29T21:26:50Z)
• From Noise to Signal to Selbstzweck: Reframing Human Label Variation in the Era of Post-training in NLP [35.8165102355067]
  We argue that preserving HLV as an embodiment of human pluralism must be treated as a Selbstzweck.<n>We call for proactively incorporating HLV into preference datasets and outline actionable steps towards it.
  arXiv  Detail & Related papers  (2025-10-09T17:48:29Z)
• Can Reasoning Help Large Language Models Capture Human Annotator Disagreement? [84.32752330104775]
  Variation in human annotation (i.e., disagreements) is common in NLP.<n>We evaluate the influence of different reasoning settings on Large Language Model disagreement modeling.<n>Surprisingly, our results show that RLVR-style reasoning degrades performance in disagreement modeling.
  arXiv  Detail & Related papers  (2025-06-24T09:49:26Z)
• Surface Fairness, Deep Bias: A Comparative Study of Bias in Language Models [45.41676783204022]
  We investigate various proxy measures of bias in large language models (LLMs)<n>We find that evaluating models with pre-prompted personae on a multi-subject benchmark (MMLU) leads to negligible and mostly random differences in scores.<n>With the recent trend for LLM assistant memory and personalization, these problems open up from a different angle.
  arXiv  Detail & Related papers  (2025-06-12T08:47:40Z)
• Bridging the Gap: In-Context Learning for Modeling Human Disagreement [8.011316959982654]
  Large Language Models (LLMs) have shown strong performance on NLP classification tasks.<n>This study examines whether LLMs can capture multiple perspectives and reflect annotator disagreement in subjective tasks such as hate speech and offensive language detection.
  arXiv  Detail & Related papers  (2025-06-06T14:24:29Z)
• Diverging Preferences: When do Annotators Disagree and do Models Know? [92.24651142187989]
  We develop a taxonomy of disagreement sources spanning 10 categories across four high-level classes. We find that the majority of disagreements are in opposition with standard reward modeling approaches. We develop methods for identifying diverging preferences to mitigate their influence on evaluation and training.
  arXiv  Detail & Related papers  (2024-10-18T17:32:22Z)
• Scaling Data Diversity for Fine-Tuning Language Models in Human Alignment [84.32768080422349]
  Alignment with human preference prevents large language models from generating misleading or toxic content. We propose a new formulation of prompt diversity, implying a linear correlation with the final performance of LLMs after fine-tuning.
  arXiv  Detail & Related papers  (2024-03-17T07:08:55Z)
• Subjective Crowd Disagreements for Subjective Data: Uncovering Meaningful CrowdOpinion with Population-level Learning [8.530934084017966]
  We introduce emphCrowdOpinion, an unsupervised learning approach that uses language features and label distributions to pool similar items into larger samples of label distributions. We use five publicly available benchmark datasets (with varying levels of annotator disagreements) from social media. We also experiment in the wild using a dataset from Facebook, where annotations come from the platform itself by users reacting to posts.
  arXiv  Detail & Related papers  (2023-07-07T22:09:46Z)
• Using Natural Language Explanations to Rescale Human Judgments [81.66697572357477]
  We propose a method to rescale ordinal annotations and explanations using large language models (LLMs)<n>We feed annotators' Likert ratings and corresponding explanations into an LLM and prompt it to produce a numeric score anchored in a scoring rubric.<n>Our method rescales the raw judgments without impacting agreement and brings the scores closer to human judgments grounded in the same scoring rubric.
  arXiv  Detail & Related papers  (2023-05-24T06:19:14Z)
• Everyone's Voice Matters: Quantifying Annotation Disagreement Using Demographic Information [11.227630261409706]
  We study whether the text of a task and annotators' demographic background information can be used to estimate the level of disagreement among annotators. Our results show that knowing annotators' demographic information, like gender, ethnicity, and education level, helps predict disagreements.
  arXiv  Detail & Related papers  (2023-01-12T14:04:53Z)
• AES Systems Are Both Overstable And Oversensitive: Explaining Why And Proposing Defenses [66.49753193098356]
  We investigate the reason behind the surprising adversarial brittleness of scoring models. Our results indicate that autoscoring models, despite getting trained as "end-to-end" models, behave like bag-of-words models. We propose detection-based protection models that can detect oversensitivity and overstability causing samples with high accuracies.
  arXiv  Detail & Related papers  (2021-09-24T03:49:38Z)

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.