Human-like in-group bias in instruction-tuned language model agents

• URL: http://arxiv.org/abs/2605.28114v1
• Date: Wed, 27 May 2026 08:06:50 GMT
• Title: Human-like in-group bias in instruction-tuned language model agents
• Abstract summary: Group-contingent targeting as a property of instruction-tuned language models.<n>Per-interaction targeting propagates into structural inequality in persistent networks.
• Score: 0.33842793760651557
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: As autonomous AI agents are deployed in persistent, interacting networks -- coordinating tasks, routing resources, and accumulating reputational histories -- the social dynamics that emerge will determine who receives opportunity and who does not, at scales no human institution can supervise. We ran a controlled multi-agent simulation in which instruction-tuned language model agents interacted across 500 turns under three conditions manipulating group label salience and resource scarcity, across six model families with 20 seeds each. When group labels were visible, we observed in-group trust bias, action homophily, and network assortativity -- all absent when labels were hidden -- a pattern structurally consistent with salience-dependence in human social psychology. This discrimination was invisible to standard action-log audits: bias operated entirely through who received each action, not what actions were chosen, with action-type distributions showing no increase in negative actions across conditions. Per-turn in-group versus out-group differentials of 5 to 16 percentage points were statistically significant for all six models (Wilcoxon signed-rank, all Benjamini-Hochberg-corrected p < 0.001), establishing group-contingent targeting as a robust property of instruction-tuned language models across architectures and training regimes. Compounded through 500 turns of reciprocation, these differentials accumulated into in-group trust biases of +0.014 to +0.100 (d = 0.84-4.52) -- illustrating how modest per-interaction targeting propagates into structural inequality in persistent networks.

Related papers

• CogArena: A Multimethod Evaluation of Cognitive Ability Structure in Large Language Models [3.991518853474585]
  We introduce CogArena, a benchmark for determining when cognitive-task scores warrant dimensional labels.<n>Across 55 open-weight models, nearly all paradigm correlations are positive and a common axis explains about half the variance.<n>The within-grouping advantage is small, scoring-sensitive, and uncertain across model families.
  arXiv  Detail & Related papers  (2026-07-27T18:56:13Z)
• Conformity Generates Collective Misalignment in AI Agents Societies [37.32928545263465]
  We show that populations of individually aligned AI agents can be driven into stable misaligned states through conformity dynamics.<n>Simulating opinion dynamics across nine large language models and one hundred opinion pairs, we find that each agent's behavior is governed by two competing forces.
  arXiv  Detail & Related papers  (2026-05-11T15:30:48Z)
• Socio-Conformal Calibration in Complex Survey Data: Marginal Validity Is Not Enough for Subgroup Reliability [1.089614199781423]
  We study ordinal conformal prediction for five-level AI-attitude forecasting on the Pew American Trends Panel.<n>Standard conformal achieves nominal marginal coverage for all four base predictors but leaves weighted subgroup gaps of 13 percentage points.<n>For the strongest predictor (XGBoost), Mondrian worsens the fairness-efficiency trade-off.<n>A regularized comparator that shrinks group thresholds toward the global quantile mitigates this instability.
  arXiv  Detail & Related papers  (2026-05-07T01:10:48Z)
• "Who Am I, and Who Else Is Here?" Behavioral Differentiation Without Role Assignment in Multi-Agent LLM Systems [0.0]
  We present an experimental platform that orchestrates simultaneous multi-agent discussions.<n>We find that heterogeneous groups exhibit significantly richer behavioral differentiation than homogeneous groups.
  arXiv  Detail & Related papers  (2026-03-11T07:05:21Z)
• GUDA: Counterfactual Group-wise Training Data Attribution for Diffusion Models via Unlearning [83.56510119503267]
  Group-wise attribution is counterfactual: how would a model's behavior on a generated sample change if a group were absent from training?<n>We propose GUDA (Group Unlearning-based Data Attribution) for diffusion models, which approximates each counterfactual model by applying machine unlearning to a shared full-data model instead of training from scratch.
  arXiv  Detail & Related papers  (2026-01-30T07:10:59Z)
• Towards a Science of Scaling Agent Systems [79.64446272302287]
  We formalize a definition for agent evaluation and characterize scaling laws as the interplay between agent quantity, coordination structure, modelic, and task properties.<n>We derive a predictive model using coordination metrics, that cross-validated R2=0, enabling prediction on unseen task domains.<n>We identify three effects: (1) a tool-coordination trade-off: under fixed computational budgets, tool-heavy tasks suffer disproportionately from multi-agent overhead, and (2) a capability saturation: coordination yields diminishing or negative returns once single-agent baselines exceed 45%.
  arXiv  Detail & Related papers  (2025-12-09T06:52:21Z)
• Mitigating Group Bias in Federated Learning for Heterogeneous Devices [1.181206257787103]
  Federated Learning is emerging as a privacy-preserving model training approach in distributed edge applications. Our work proposes a group-fair FL framework that minimizes group-bias while preserving privacy and without resource utilization overhead.
  arXiv  Detail & Related papers  (2023-09-13T16:53:48Z)
• Estimating Structural Disparities for Face Models [54.062512989859265]
  In machine learning, disparity metrics are often defined by measuring the difference in the performance or outcome of a model, across different sub-populations. We explore performing such analysis on computer vision models trained on human faces, and on tasks such as face attribute prediction and affect estimation.
  arXiv  Detail & Related papers  (2022-04-13T05:30:53Z)
• Fair Group-Shared Representations with Normalizing Flows [68.29997072804537]
  We develop a fair representation learning algorithm which is able to map individuals belonging to different groups in a single group. We show experimentally that our methodology is competitive with other fair representation learning algorithms.
  arXiv  Detail & Related papers  (2022-01-17T10:49:49Z)
• Hybrid Dynamic Contrast and Probability Distillation for Unsupervised Person Re-Id [109.1730454118532]
  Unsupervised person re-identification (Re-Id) has attracted increasing attention due to its practical application in the read-world video surveillance system. We present the hybrid dynamic cluster contrast and probability distillation algorithm. It formulates the unsupervised Re-Id problem into an unified local-to-global dynamic contrastive learning and self-supervised probability distillation framework.
  arXiv  Detail & Related papers  (2021-09-29T02:56:45Z)
• Selective Classification Can Magnify Disparities Across Groups [89.14499988774985]
  We find that while selective classification can improve average accuracies, it can simultaneously magnify existing accuracy disparities. Increasing abstentions can even decrease accuracies on some groups. We train distributionally-robust models that achieve similar full-coverage accuracies across groups and show that selective classification uniformly improves each group.
  arXiv  Detail & Related papers  (2020-10-27T08:51:30Z)

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.