🤖 AI Summary
Manipulative behavior detection in courtroom dialogues has long suffered from a lack of systematic investigation and high-quality, annotated data. To address this gap, we propose CLAIM—a novel, intent-driven, two-stage multi-agent framework for manipulative behavior detection and attribution in judicial settings. Our method integrates dialogue state tracking, legal-domain fine-tuned language models, fine-grained sequence labeling, and relation extraction. We introduce LegalCon, the first large-scale, long-context, expert-annotated courtroom dialogue dataset (1,063 trials), with explicit annotations of manipulators, manipulation tactics, and context-dependent relational dependencies. On LegalCon, CLAIM achieves 82.3% F1 for manipulation detection and 79.6% accuracy for manipulator identification—significantly outperforming single-model baselines. We publicly release both code and data, establishing the first benchmark for modeling manipulative discourse in legal NLP and advancing explainable, justice-oriented AI systems.
📝 Abstract
Courtrooms are places where lives are determined and fates are sealed, yet they are not impervious to manipulation. Strategic use of manipulation in legal jargon can sway the opinions of judges and affect the decisions. Despite the growing advancements in NLP, its application in detecting and analyzing manipulation within the legal domain remains largely unexplored. Our work addresses this gap by introducing LegalCon, a dataset of 1,063 annotated courtroom conversations labeled for manipulation detection, identification of primary manipulators, and classification of manipulative techniques, with a focus on long conversations. Furthermore, we propose CLAIM, a two-stage, Intent-driven Multi-agent framework designed to enhance manipulation analysis by enabling context-aware and informed decision-making. Our results highlight the potential of incorporating agentic frameworks to improve fairness and transparency in judicial processes. We hope that this contributes to the broader application of NLP in legal discourse analysis and the development of robust tools to support fairness in legal decision-making. Our code and data are available at https://github.com/Disha1001/CLAIM.