ACL Findings2024
Knowledge-Infused Legal Wisdom: Navigating LLM Consultation through the Lens of Diagnostics and Positive-Unlabeled Reinforcement Learning
Yang Wu, Chenghao Wang, Ece Gumusel, Xiaozhong Liu
Abstract
The integration of generative Large Language Models (LLMs) into various applications, including the legal domain, has been accelerated by their expansive and versatile nature. However, when facing a legal case, users without a legal background often struggle to formulate professional queries and may inadvertently overlook critical legal factors when presenting their case narrative to LLMs. To address this issue, we propose the Diagnostic Legal Large Language Model (D3LM), which utilizes adaptive lawyer-like diagnostic questions to collect additional case information and then provides high-quality feedback. D3LM incorporates an innovative graph-based Positive-Unlabeled Reinforcement Learning (PURL) algorithm, enabling the generation of critical questions and enhancing user-LLM interactions. Moreover, an integrated LLM-based stopping criterion facilitates precise Court Views Generation (CVG). Our research also introduces a new English-language CVG dataset 1 based on the US case law database, enriching the realm of LLM research and deployment with a vital dimension. D3LM surpasses classical LLMs by delivering outstanding performance and a remarkable user experience in the legal domain.
Ask Lune
Know this paper in minutes, not an afternoon.
Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.
1 top-tier paper has cited it since. Know what they found before your reviewers do.
Lune can draw on
the full paper · the top-tier papers that cite it · related work
Free to start. No credit card required.
Lineage
Cited by top-tier papers1
Builds on13
- Toolformer: Language Models Can Teach Themselves to Use ToolsTimo Schick, Jane Dwivedi-Yu, Roberto Dessì, Roberta Raileanu et al.NeurIPS 2023 · 5,989 citations
- Why Johnny Can't Prompt: How Non-AI Experts Try (and Fail) to Design LLM PromptsJ. D. Zamfirescu-Pereira, Richmond Y. Wong, Bjoern Hartmann, Qian YangCHI 2023 · 892 citations
- Neural Contextual Bandits with UCB-based ExplorationDongruo Zhou, Lihong Li, Quanquan GuICML 2020 · 329 citations
- LLM Maybe LongLM: SelfExtend LLM Context Window Without TuningHongye Jin, Xiaotian Han, Jingfeng Yang, Zhimeng Jiang et al.ICML 2024 · 167 citations
- Digraph Inception Convolutional NetworksZekun Tong, Yuxuan Liang, Changsheng Sun, Xinke Li et al.NeurIPS 2020 · 132 citations
Related papers
- Generating Novel Leads for Drug Discovery Using LLMs with Logical FeedbackShreyas Bhat Brahmavar, Ashwin Srinivasan, Tirtharaj Dash, Sowmya Ramaswamy Krishnan et al.AAAI 2024 · 23 citations
- Leveraging Large Language Models for Learning Complex Legal Concepts through StorytellingHang Jiang, Xiajie Zhang, Robert Mahari, Daniel T. Kessler et al.ACL 2024 · 12 citations
- Language Agents for Hypothesis-driven Clinical Decision Making with Reinforcement LearningDavid Bani-Harouni, Chantal Pellegrini, Ege Özsoy, Nassir Navab et al.ICLR 2026 · 16 citations
- Diagnostic-Guided Dynamic Profile Optimization for LLM-based User Simulators in Sequential RecommendationHongyang Liu, Zhu Sun, Tianjun Wei, Yan Wang et al.AAAI 2026 · 4 citations
- DiagLink: A Dual-User Diagnostic Assistance System by Synergizing Experts with LLMs and Knowledge GraphsZihan Zhou, Yinan Liu, Yuyang Xie, Bin Wang et al.CHI 2026 · 1 citation
