SAILER: Structure-aware Pre-trained Language Model for Legal Case Retrieval
Open accessAbstract
Legal case retrieval, which aims to find relevant cases for a query case, plays a core role in the intelligent legal system. Despite the success that pre-training has achieved in ad-hoc retrieval tasks, effective pre-training strategies for legal case retrieval remain to be explored. Compared with general documents, legal case documents are typically long text sequences with intrinsic logical structures. However, most existing language models have difficulty understanding the long-distance dependencies between different structures. Moreover, in contrast to the general retrieval, the relevance in the legal domain is sensitive to key legal elements. Even subtle differences in key legal elements can significantly affect the judgement of relevance. However, existing pre-trained language models designed for general purposes have not been equipped to handle legal elements.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Building an actual LM for corpus modeling enables capturing implicit domain knowledge and potentially saves tokens in prompts.
“building an actual LM for corpus modeling enable us to capture implicit domain knowledge (e.g., the fine-grained differences between law articles [152]) and potentially save tokens in prompts.”
From Foundations of GenIR (2025) · cites this paper as [152]
Cites
Optimizing Dense Retrieval Model Training with Hard Negatives
Zhan, J., Mao, J., Liu, Y., Guo, J., Zhang, M., & Ma, S. (2021) · arXiv · doi:10.48550/arxiv.2104.08051 · Cited by 251 · see citing works
Self-citationOptimizing Dense Retrieval Model Training with Hard Negatives
Zhan, J., Mao, J., Liu, Y., Guo, J., Zhang, M., & Ma, S. (2021) · Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval · doi:10.1145/3404835.3462880 · Cited by 248 · see citing works
Self-citation
Flagged first, then by citations
Cited by
Sorted by citations
Ai, Q., Zhan, J., & Liu, Y. (2025) · arXiv · doi:10.48550/arxiv.2501.02842
“…the overall idea of prompt general LLMs with domain-specific prompts is similar to the framework of RAG, building an actual LM for corpus modeling enable us to capture implicit domain knowledge (e.g., the fine-grained differences between law articles [152]) and potentially save tokens in prompts.”
Self-citation
- Authors
- Li, H., Ai, Q., Chen, J., Dong, Q., Wu, Y., Liu, Y., Chen, C., & Tian, Q.
- Affiliations
- Tsinghua University (CN)
- Huawei Technologies (China) (CN)
- Venue
- Proceedings of the 46th International ACM SIGIR Conference on Research and Development in Information Retrieval
- Publisher
- ACM
- Published
- 2023-07-18
- Pages
- 1035-1044
- OpenAlex
- W4366999416
- Cited by
- 79
- References
- 2
- Citations by year
- 2026: 32 · 2025: 22 · 2024: 19 · 2023: 6
- FWCI
- 107.28
- Type
- conference paper
- Language
- en
- Open access
- green
- arxiv.org
- Locations
- International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR) (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University)
- Funders
- Tsinghua University Guoqiang Research Institute
- Natural Science Foundation of China (62002194)
- Tsinghua-Tencent Tiangong Institute for Intelligent Computing
- Topics
- Social Sciences › Political Science and International Relations › Artificial Intelligence in Law
- Physical Sciences › Computer Science › Information Systems › Information Retrieval and Search Behavior
- Physical Sciences › Computer Science › Artificial Intelligence › Topic Modeling
- Keywords
- case law searching, pre-trained language models, long-distance dependencies, legal elements, relevance assessment, legal text understanding
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
