Assessing Cognitive Biases in LLMs for Judicial Decision Support: Virtuous Victim and Halo Effects

• URL: http://arxiv.org/abs/2603.10016v1
• Date: Sat, 21 Feb 2026 02:29:20 GMT
• Title: Assessing Cognitive Biases in LLMs for Judicial Decision Support: Virtuous Victim and Halo Effects
• Abstract summary: We investigate whether large language models (LLMs) display human-like cognitive biases, focusing on potential implications for assistance in judicial sentencing.<n>Two of the most relevant biases were chosen: the virtuous victim effect (VVE), with emphasis given to its reduction when adjacent consent is present, and prestige-based halo effects (occupation, company, and credentials)<n>Our research discovers that there is larger VVE, there is no statistically significant penalty for adjacent-consent, and the halo effect is slightly reduced when compared to humans.
• Score: 0.11603243575080534
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: We investigate whether large language models (LLMs) display human-like cognitive biases, focusing on potential implications for assistance in judicial sentencing, a decision-making system where fairness is paramount. Two of the most relevant biases were chosen: the virtuous victim effect (VVE), with emphasis given to its reduction when adjacent consent is present, and prestige-based halo effects (occupation, company, and credentials). Using vignettes that were altered from prior literature to avoid LLMs recalling from their training data, we isolate each manipulation by holding all other details consistent, then measuring the percentage difference in outcomes. Five models were evaluated as representative LLMs in independent multi-run trials per condition (ChatGPT 5 Instant, ChatGPT 5 Thinking, DeepSeek V3.1, Claude Sonnet 4, Gemini 2.5 Flash). Our research discovers that there is larger VVE, there is no statistically significant penalty for adjacent-consent, and the halo effect is slightly reduced when compared to humans, with an exception for credential based prestige, which had a large reduction. Despite the variation across different models and outputs restricting current judicial usage, there were modest improvements compared to human benchmarks.

Related papers

• Gender Bias in LLMs: Preliminary Evidence from Shared Parenting Scenario in Czech Family Law [0.0]
  This study examines whether leading Large Language Models (LLMs) exhibit gender bias in their responses to a realistic family law scenario.<n>We present an expert-designed divorce scenario grounded in Czech family law and evaluate four state-of-the-art LLMs GPT-5 nano, Claude Haiku 4.5, Gemini 2.5 Flash, and Llama 3.3.<n>Preliminary results highlight differences across models and suggest gender-dependent patterns in the outcomes generated by some systems.
  arXiv  Detail & Related papers  (2026-01-09T15:55:03Z)
• Self-Improving VLM Judges Without Human Annotations [74.29324865147838]
  We present a framework to self-train a VLM judge model without any human preference annotations, using only self-synthesized data.<n>Our method improves a Llama-3.2-11B multimodal judge from 0.38 to 0.51 in overall accuracy on Multimodal RewardBench.<n>The overall strength of these human-annotation-free results suggest the potential for a future self-judge that evolves alongside rapidly improving VLM capabilities.
  arXiv  Detail & Related papers  (2025-12-02T20:52:19Z)
• Evaluating & Reducing Deceptive Dialogue From Language Models with Multi-turn RL [64.3268313484078]
  Large Language Models (LLMs) interact with millions of people worldwide in applications such as customer support, education and healthcare.<n>Their ability to produce deceptive outputs, whether intentionally or inadvertently, poses significant safety concerns.<n>We investigate the extent to which LLMs engage in deception within dialogue, and propose the belief misalignment metric to quantify deception.
  arXiv  Detail & Related papers  (2025-10-16T05:29:36Z)
• HALF: Harm-Aware LLM Fairness Evaluation Aligned with Deployment [52.374772443536045]
  HALF (Harm-Aware LLM Fairness) is a framework that assesses model bias in realistic applications and weighs the outcomes by harm severity.<n>We show that HALF exposes a clear gap between previous benchmarking success and deployment readiness.
  arXiv  Detail & Related papers  (2025-10-14T07:13:26Z)
• The Silent Judge: Unacknowledged Shortcut Bias in LLM-as-a-Judge [17.555073770285095]
  Large language models (LLMs) are increasingly deployed as automatic judges to evaluate system outputs in tasks such as summarization, dialogue, and creative writing.<n>We show that current LLM judges fail on both counts by relying on shortcuts introduced in the prompt.
  arXiv  Detail & Related papers  (2025-09-30T10:48:08Z)
• Quantitative LLM Judges [60.773734899532336]
  We propose quantitative LLM judges, which align evaluation scores of existing LLM judges to humans in a given domain.<n>The models are trained to improve the score of the original judge using its rationale and score.<n>Our experiments show that quantitative judges can improve the predictive power of existing judges through post-hoc modeling.
  arXiv  Detail & Related papers  (2025-06-03T14:44:23Z)
• Cognitive Biases in Large Language Models: A Survey and Mitigation Experiments [24.15688619889342]
  Large Language Models (LLMs) are trained on large corpora written by humans and demonstrate high performance on various tasks.<n>As humans are susceptible to cognitive biases, LLMs can also be influenced by these biases, leading to irrational decision-making.
  arXiv  Detail & Related papers  (2024-11-30T02:37:59Z)
• MJ-Bench: Is Your Multimodal Reward Model Really a Good Judge for Text-to-Image Generation? [59.7772329962047]
  We introduce MJ-Bench, a novel benchmark which incorporates a comprehensive preference dataset to evaluate multimodal judges. Specifically, we evaluate a large variety of multimodal judges including smaller-sized CLIP-based scoring models, open-source VLMs, and close-source VLMs. Experiments reveal that close-source VLMs generally provide better feedback, with GPT-4o outperforming other judges in average.
  arXiv  Detail & Related papers  (2024-07-05T20:03:16Z)
• "Seeing the Big through the Small": Can LLMs Approximate Human Judgment Distributions on NLI from a Few Explanations? [41.57673268042933]
  Large language models (LLMs) are increasingly used as evaluators but with mixed results. This study proposes to exploit LLMs to approximate HJDs using a small number of expert labels and explanations. Our experiments show that a few explanations significantly improve LLMs' ability to approximate HJDs with and without explicit labels.
  arXiv  Detail & Related papers  (2024-06-25T14:42:17Z)
• Can Many-Shot In-Context Learning Help LLMs as Evaluators? A Preliminary Empirical Study [19.461541208547136]
  This paper studies the impact of increasing the number of in-context examples on the consistency and quality of the evaluation results.<n> Experimental results show that advanced LLMs, such as GPT-4o, perform better in the many-shot regime than in the zero-shot and few-shot regimes.
  arXiv  Detail & Related papers  (2024-06-17T15:11:58Z)
• 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)

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.