Modeling Motivated Reasoning in Law: Evaluating Strategic Role Conditioning in LLM Summarization

2025/08/30 by Eunjung Cho, Alexander Hoyle, Cho, Eunjung +3 · 1 citation
Computer Science · Social Sciences · #Artificial Intelligence in Law #Computation and Language (cs.CL) #Computers and Society (cs.CY) #FOS: Computer and information sciences #Multi-Agent Systems and Negotiation #Topic Modeling

paper · pdf · doi:10.48550/arxiv.2509.00529

openalex publication_date 2025/08/30 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Large Language Models (LLMs) are increasingly used to generate user-tailored summaries, adapting outputs to specific stakeholders. In legal contexts, this raises important questions about motivated reasoning -- how models strategically frame information to align with a stakeholder's position within the legal system. Building on theories of legal realism and recent trends in legal practice, we investigate how LLMs respond to prompts conditioned on different legal roles (e.g., judges, prosecutors, attorneys) when summarizing judicial decisions. We introduce an evaluation framework grounded in legal fact and reasoning inclusion, also considering favorability towards stakeholders. Our results show that even when prompts include balancing instructions, models exhibit selective inclusion patterns that reflect role-consistent perspectives. These findings raise broader concerns about how similar alignment may emerge as LLMs begin to infer user roles from prior interactions or context, even without explicit role instructions. Our results underscore the need for role-aware evaluation of LLM summarization behavior in high-stakes legal settings.

Citations

• The Inadequacy of Offline LLM Evaluations: A Need to Account for Personalization in Model Behavior
• Neither Valid nor Reliable? Investigating the Use of LLMs as Judges
• Values in the Wild: Discovering and Analyzing Values in Real-World Language Model Interactions
• REALM: A Dataset of Real-World LLM Use Cases
• NoLiMa: Long-Context Evaluation Beyond Literal Matching
• The Alternative Annotator Test for LLM-as-a-Judge: How to Statistically Justify Replacing Human Annotators with LLMs
• Limitations of the LLM-as-a-Judge Approach for Evaluating LLM Outputs in Expert Knowledge Tasks
• Unlocking Legal Knowledge: A Multilingual Dataset for Judicial Summarization in Switzerland
• FineSurE: Fine-grained Summarization Evaluation using LLMs
• SafetyPrompts: a Systematic Review of Open Datasets for Evaluating and Improving Large Language Model Safety
• LexAbSumm: Aspect-based Summarization of Legal Decisions
• Quantifying the Persona Effect in LLM Simulations
• Promises and pitfalls of artificial intelligence for legal applications
• Toward a General Framework of Biased Reasoning: Coherence-Based Reasoning
• AI Supported Degradation of the Self Concept: A Theoretical Framework Grounded in Established Cognitive and Computational Mechanisms
• Open Problems and Fundamental Limitations of Reinforcement Learning from Human Feedback
• Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena
• Rethinking Model Evaluation as Narrowing the Socio-Technical Gap
• G-Eval: NLG Evaluation using GPT-4 with Better Human Alignment
• News Summarization and Evaluation in the Era of GPT-3
• Using Large Language Models to Simulate Multiple Humans and Replicate Human Subject Studies
• Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback
• Training language models to follow instructions with human feedback
• Re-evaluating Evaluation in Text Summarization
• Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks
• The case for motivated reasoning.

Cited by

• LAMUS: A Large-Scale Corpus for Legal Argument Mining from U.S. Caselaw using LLMs

Related

• Structured Definitions and Segmentations for Legal Reasoning in LLMs: A Study on Indian Legal Data
• Thinking Longer, Not Always Smarter: Evaluating LLM Capabilities in Hierarchical Legal Reasoning
• LegalBench: A Collaboratively Built Benchmark for Measuring Legal Reasoning in Large Language Models
• Do LLMs Truly “Understand” when a Precedent Is Overruled?
• Can Large Language Models Grasp Legal Theories? Enhance Legal Reasoning with Insights from Multi-Agent Collaboration