计算机科学 ›› 2025, Vol. 52 ›› Issue (8): 308-316.doi: 10.11896/jsjkx.240900170
王东升
WANG Dongsheng
摘要： 一些研究利用先进的大模型(LLM)技术理解法律事实,预测被告人的罪名、刑期等判决结果。为进一步深入研究,选择了更为复杂的多被告人法律判决预测任务,它比单被告人预测更困难。具体地,将与LLM的交互由单轮升级为多轮,以此提高LLM对案件的理解能力。此外,构建了描述案件的两类犯罪知识图谱,其中犯罪关系知识图谱刻画了被告人之间的帮助关系,量刑情节知识图谱展示了案件的核心犯罪情节。通过犯罪知识图谱,设计了一个检索系统为LLM提供类案判决的参考。在多被告法律判决预测实验中,所提方法的预测结果优于对比方法,这表明多轮LLM交互和犯罪知识图谱的设计是有效的。
中图分类号:
| [1]TONG S X,YUAN J L,ZHANG P L,et al.Legal Judgment Prediction via graph boosting with constraints[J].Information Processing and Management,2024,61(2):103663. [2]ZHANG C Y,QU H,CUI C R,et al.Process Supervision Based Sequence Multi-task Method for Legal Judgement Prediction[J].Computer Science,2021,48(3):227-232. [3]GAN L L,LI B K,KUANG K,et al.Exploiting Contrastive Learning and Numerical Evidence for Confusing Legal Judgment Prediction[C]//Conference on Empirical Methods in Natural Language Processing.2023:12174-12185. [4]FENG G Y,QIN Y B,HUANG R Z,et al.Criminal Action Graph:A semantic representation model of judgement documents for legal charge prediction[J].Information Processing and Management,2024,60(5):103421. [5]SUKANYA G,JAYARAJU P.Modified Hierarchical-Attention Network model for legal judgment predictions[J].Data & Knowledge Engineering,2023,147:102203. [6]ZHAO Q H,GAO T H,GUO N.LA-MGFM:A legal judgment prediction method via sememe-enhanced graph neural networks and multi-graph fusion mechanism[J].Information Processing and Management,2023,60(5):103455. [7]XIAO C J,HU X Y,LIU Z Y,et al.Lawformer:A pre-trained language model for Chinese legal long documents[J].AI Open,2021,2:79-84. [8]ZHOU Z,SHI J X,SONG P X,et al.LawGPT:A Chinese Legal Knowledge-Enhanced Large Language Model[J].arXiv:2406.04614,2024. [9]PAN S P,LU T,GU N,et al.Charge Prediction for Multi-defendant Cases with Multi-scale Attention[C]//Computer Supported Cooperative Work and Social Computing.2019:766-777. [10]LYU Y G,HAO J T,WANG Z H,et al.Multi-Defendant Legal Judgment Prediction via Hierarchical Reasoning[C]//Findings of the Association for Computational Linguistics.2023:2198-2209. [11]QIN W C,CAO Z L,YU W J,et al.Explicitly Integrating Judgment Prediction with Legal Document Retrieval:A Law-Guided Generative Approach[C]//The 47th International ACM SIGIR Conference on Research and Development in Information Retrieval.2024:2210-2220. [12]KORT F.Predicting Supreme Court Decisions Mathematically:A Quantitative Analysis of the “Right to Counsel” Cases[J].The American Political Science Review,1957,51(1):1-12. [13]FENG Y,LI C Y,NG V.Legal Judgment Prediction via Event Extraction with Constraints[C]//Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics.2022:648-664. [14]LIU Y F,WU Y Q,ZHANG Y T,et al.ML-LJP:Multi-LawAware Legal Judgment Prediction[C]//The 46th International ACM SIGIR Conference on Research and Development in Information Retrieval.2023:1023-1034. [15]CHALKIDIS I,FERGADIOTIS M,MALAKASIOTIS P,et al.LEGAL-BERT:The Muppets straight out of Law School[J].arXiv:2010.02559,2020. [16]BELTAGY L,PETERS M,COHAN A.Longformer:The Long-Document Transformer[J].arXiv:2004.05150,2020. [17]OUYANG L,WU J,JIANG X,et al.Training language models to follow instructions with human feedback[C]//Annual Conference on Neural Information Processing Systems 2022.2022:27730-27744. [18]SUN Q,HUANG K,YANG X C,et al.Consistency GuidedKnowledge Retrieval and Denoising in LLMs for Zero-shot Do-cument-level Relation Triplet Extraction[C]//The ACM on Web Conference.2024:4407-4416. [19]JI B J,DUAN X Y,QIU Z Y,et al.Submodular-based In-context Example Selection for LLMs-based Machine Translation[C]//The 2024 Joint International Conference on ComputationalLinguistics,Language Resources and Evaluation.2024:15398-15409. [20]AGRAWAL G,PAL K,DENG Y L,et al.CyberQ:Generating Questions and Answers for Cybersecurity Education Using Knowledge Graph-Augmented LLMs[C]//Thirty-Eighth AAAI Conference on Artificial Intelligence.2024:23164-23172. [21]HUANG Q Z,TAO M X,AN Z W,et al.Lawyer LLaMA Tech-nical Report[J].arXiv:2305.15062,2023. [22]WU Y Q,ZHOU S Y,LIU Y F,et al.Precedent-Enhanced Legal Judgment Prediction with LLM and Domain-Model Collaboration[C]//The 2023 Conference on Empirical Methods in Natural Language Processing.2023:12060-12075. [23]SHUI R H,CAO Y X,WANG X,et al.A Comprehensive Evaluation of Large Language Models on Legal Judgment Prediction[C]//Findings of the Association for Computational Linguistics:EMNLP 2023.2023:7337-7348. [24]DENG C L,MAO K L,ZHANG Y Y,et al.Enabling Discriminative Reasoning in LLMs for Legal Judgment Prediction[J].arXiv:2407.01964,2024. [25]WEI J,WANG X Z,SCHUURMANS D,et al.Chain-of-Thought Prompting Elicits Reasoning in Large Language Models[C]//Annual Conference on Neural Information Processing Systems.2022:24824-24837. [26]ZHANG Z P,WANG A Q,ZHANG Y,et al.Relation pruning and discriminative sampling over knowledge graph for long-tail recommendation[J].Information Sciences,2024,679:120871. [27]LIU H Z,ZHU Y,WU Z H.Knowledge Graph-Based Behavior Denoising and Preference Learning for Sequential Recommendation[J].IEEE Transactions on Knowledge and Data Enginee-ring,2024,36(6):2490-2503. [28]HU Y L,ZOU F T,HAN J J,et al.LLM-TIKG:Threat intelligence knowledge graph construction utilizing large language model[J].Computers & Security,2024,145:103999. [29]FAN Z L,CHEN C C.CuPe-KG:Cultural perspective-basedknowledge graph construction of tourism resources via pretrained language models[J].Information Processing and Ma-nagement,2024,61(2):103646. [30]ZHANG Y X,ZHAI S L,MENG Y,et al.Event is more valuable than you think:Improving the Similar Legal Case Retrieval via event knowledge[J].Information Processing and Management,2024,61(4):103729. [31]CORREIA F,ALMEIDA A,et al.Fine-grained legal entity annotation:A case study on the Brazilian Supreme Court[J].Information Processing and Management,2022,59(1):102794. [32]LIU D G,DU W H,LI L,et al.Augmenting Legal JudgmentPrediction with Contrastive Case Relations[C]//The 29th International Conference on Computational Linguistics.2022:2658-2667. [33]YANG W M,JIA W J,ZHOU X J,et al.Legal Judgment Prediction via Multi-Perspective Bi-Feedback Network[C]//The Twenty-Eighth International Joint Conference on Artificial Intelligence.2019:4085-4091. [34]XU N,WANG P H,CHEN L,et al.Distinguish Confusing Law Articles for Legal Judgment Prediction[C]//The 58th Annual Meeting of the Association for Computational Linguistics.2020:3086-3095. [35]YUE L N,LIU Q,JIN B B,et al.NeurJudge:A Circumstance-aware Neural Framework for Legal Judgment Prediction[C]//The 44th International ACM SIGIR Conference on Research and Development in Information Retrieval.2021:973-982. [36]CUI Y M,CHE W X,LIU T,et al.Pre-Training With Whole Word Masking for Chinese BERT[J].IEEE/ACM Transactions on Audio,Speech and Language Processing,2021,29:3504-3514. [37]XUE L T,CONSTANT N,ROBERTS A,et al.mT5:A Massively Multilingual Pre-trained Text-to-Text Transformer[C]//The 2021 Conference of the North American Chapter of the Association for Computational Linguistics:Human Language Technologies.2021:483-498. |
|
||
首页