LLM Agents in Law: Taxonomy, Applications, and Challenges
2026/01/08 by Shuang Liu, Ruijia Zhang, Ruoyun Ma +6 · 1 voice
Computer Science · #cs.AI #cs.CY
paper · pdf · doi:10.48550/arxiv.2601.06216
Abstract
Large language models (LLMs) have precipitated a dramatic improvement in the legal domain, yet the deployment of standalone models faces significant limitations regarding hallucination, outdated information, and verifiability. Recently, LLM agents have attracted significant attention as a solution to these challenges, utilizing advanced capabilities such as planning, memory, and tool usage to meet the rigorous standards of legal practice. In this paper, we present a comprehensive survey of LLM agents for legal tasks, analyzing how these architectures bridge the gap between technical capabilities and domain-specific needs. Our major contributions include: (1) systematically analyzing the technical transition from standard legal LLMs to legal agents; (2) presenting a structured taxonomy of current agent applications across distinct legal practice areas; (3) discussing evaluation methodologies specifically for agentic performance in law; and (4) identifying open challenges and outlining future directions for developing robust and autonomous legal assistants.
Citations
- ARCADIA: Scalable Causal Discovery for Corporate Bankruptcy Analysis Using Agentic AI
- Towards Trustworthy Legal AI through LLM Agents and Formal Reasoning
- Multi-Agent Legal Verifier Systems for Data Transfer Planning
- Building from Scratch: A Multi-Agent Framework with Human-in-the-Loop for Multilingual Legal Terminology Mapping
- Law in Silico: Simulating Legal Society with LLM-Based Agents
- Human-Centered LLM-Agent System for Detecting Anomalous Digital Asset Transactions
- Deterministic Legal Agents: A Canonical Primitive API for Auditable Reasoning over Temporal Knowledge Graphs
- NegotiationGym: Self-Optimizing Agents in a Multi-Agent Social Simulation Environment
- LegalSim: Multi-Agent Simulation of Legal Systems for Discovering Procedural Exploits
- MASLegalBench: Benchmarking Multi-Agent Systems in Deductive Legal Reasoning
- AutoSpec: An Agentic Framework for Automatically Drafting Patent Specification
- Agentic AI for Financial Crime Compliance
- An LLM Agentic Approach for Legal-Critical Software: A Case Study for Tax Prep Software
- Agent4FaceForgery: Multi-Agent LLM Framework for Realistic Face Forgery Detection
- JustEva: A Toolkit to Evaluate LLM Fairness in Legal Knowledge Inference
- Large Language Models Meet Legal Artificial Intelligence: A Survey
- Co-Investigator AI: The Rise of Agentic AI for Smarter, Trustworthy AML Compliance Narratives
- Simulating Dispute Mediation with LLM-Based Agents for Legal Research
- SAMVAD: A Multi-Agent System for Simulating Judicial Deliberation Dynamics in India
- LLM-empowered Agents Simulation Framework for Scenario Generation in Service Ecosystem Governance
- L-MARS: Legal Multi-Agent System with Agentic Search and Citation-Faithfulness Audit
- On Verifiable Legal Reasoning: A Multi-Agent Framework with Formalized Knowledge Representations
- Chinese Court Simulation with LLM-Based Agent System
- LLM-Based Agents for Competitive Landscape Mapping in Drug Asset Due Diligence
- RAGulating Compliance: A Multi-Agent Knowledge Graph for Regulatory QA
- Retrieval-Augmented Multi-Agent System for Rapid Statement of Work Generation
- ContractEval: Benchmarking LLMs for Clause-Level Legal Risk Identification in Commercial Contracts
- MASCA: LLM based-Multi Agents System for Credit Assessment
- Compliance Brain Assistant: Conversational Agentic AI for Assisting Compliance Tasks in Enterprise Environments
- Privacy Artifact ConnecTor (PACT): Embedding Enterprise Artifacts for Compliance AI Agents
- TRIDENT: Benchmarking LLM Safety in Finance, Medicine, and Law
- Ready Jurist One: Benchmarking Language Agents for Legal Intelligence in Dynamic Environments
- Legal text summarization via judicial syllogism with large language models
- PhishDebate: An LLM-Based Multi-Agent Framework for Phishing Website Detection
- Mitigating Manipulation and Enhancing Persuasion: A Reflective Multi-Agent Approach for Legal Argument Generation
- PAKTON: A Multi-Agent Framework for Question Answering in Long Legal Agreements
- LegalEval-Q: A New Benchmark for The Quality Evaluation of LLM-Generated Legal Text
- SafeScientist: Toward Risk-Aware Scientific Discoveries by LLM Agents
- AI Agent Governance: A Field Guide
- Is Your LLM-Based Multi-Agent a Reliable Real-World Planner? Exploring Fraud Detection in Travel Planning
- Enhancing LLM-Based Agents via Global Planning and Hierarchical Execution
- Debate-Feedback: A Multi-Agent Framework for Efficient Legal Judgment Prediction
- Debate-Driven Multi-Agent LLMs for Phishing Email Detection
- Multi-Agent Simulator Drives Language Models for Legal Intensive Interaction
- A Survey on Responsible LLMs: Inherent Risk, Malicious Use, and Mitigation Strategy
- LegalAgentBench: Evaluating LLM Agents in Legal Domain
- CitaLaw: Enhancing LLM with Citations in Legal Domain
- LAW: Legal Agentic Workflows for Custody and Fund Services Contracts
- AutoPatent: A Multi-Agent Framework for Automatic Patent Generation
- Political Actor Agent: Simulating Legislative System for Roll Call Votes Prediction with Large Language Models
- Can Large Language Models Grasp Legal Theories? Enhance Legal Reasoning with Insights from Multi-Agent Collaboration
- LexEval: A Comprehensive Chinese Legal Benchmark for Evaluating Large Language Models
- Towards Automated Patent Workflows: AI-Orchestrated Multi-Agent Framework for Intellectual Property Management and Analysis
- Safeguarding Decentralized Social Media: LLM Agents for Automating Community Rule Compliance
- Constructing Domain-Specific Evaluation Sets for LLM-as-a-judge
- AgentCourt: Simulating Court with Adversarial Evolvable Lawyer Agents
- LawLuo: A Multi-Agent Collaborative Framework for Multi-Round Chinese Legal Consultation
- FinCon: A Synthesized LLM Multi-Agent System with Conceptual Verbal Reinforcement for Enhanced Financial Decision Making
- Simulating The U.S. Senate: An LLM-Driven Agent Approach to Modeling Legislative Behavior and Bipartisanship
- Hallucination-Free? Assessing the Reliability of Leading AI Legal Research Tools
- A Survey on Large Language Models for Critical Societal Domains: Finance, Healthcare, and Law
- Cooperate or Collapse: Emergence of Sustainable Cooperation in a Society of LLM Agents
- Evaluating AI for Law: Bridging the Gap with Open-Source Solutions
- Enhancing Anomaly Detection in Financial Markets with an LLM-based Multi-Agent Framework
- Evaluation Ethics of LLMs in Legal Domain
- AgentsCourt: Building Judicial Decision-Making Agents with Court Debate Simulation and Legal Knowledge Augmentation
- (A)I Am Not a Lawyer, But...: Engaging Legal Experts towards Responsible LLM Policies for Legal Advice
- Large Legal Fictions: Profiling Legal Hallucinations in Large Language Models
- Large Language Models in Law: A Survey
- A Survey on Hallucination in Large Language Models: Principles, Taxonomy, Challenges, and Open Questions
- A Comprehensive Evaluation of Large Language Models on Legal Judgment Prediction
- FreshLLMs: Refreshing Large Language Models with Search Engine Augmentation
- Interpretable Long-Form Legal Question Answering with Retrieval-Augmented Large Language Models
- LawBench: Benchmarking Legal Knowledge of Large Language Models
- The Rise and Potential of Large Language Model Based Agents: A Survey
- LegalBench: A Collaboratively Built Benchmark for Measuring Legal Reasoning in Large Language Models
- Lost in the Middle: How Language Models Use Long Contexts
- Chatlaw: A Multi-Agent Collaborative Legal Assistant with Knowledge Graph Enhanced Mixture-of-Experts Large Language Model
- Reflexion: Language Agents with Verbal Reinforcement Learning
- Knowledge Graphs
- The Automated but Risky Game: Modeling and Benchmarking Agent-to-Agent Negotiations and Transactions in Consumer Markets
- MultiPhishGuard: An LLM-based Multi-Agent System for Phishing Email Detection
- Advancing AI Negotiations: New Theory and Evidence from a Large-Scale Autonomous Negotiations Competition
Discussions