Cassette: Case-to-Case Structural Distillation for Efficient Legal Case Retrieval

• URL: http://arxiv.org/abs/2609.08185v1
• Date: Tue, 08 Sep 2026 03:16:20 GMT
• Title: Cassette: Case-to-Case Structural Distillation for Efficient Legal Case Retrieval
• Abstract summary: High computational cost during inference on large-scale legal databases limits its practical use in real-world settings.<n>In this paper, Cassette framework is proposed with a ranking objective and eigen-matching objective for an effective transfer of knowledge from a powerful and well-trained heavy teacher retriever to a lightweight and efficient hybrid student dual encoder.
• Score: 60.39519301815861
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Legal case retrieval (LCR) is an essential tool for not only assisting legal practitioners to efficiently retrieve precedents but also enabling ordinary individuals to find valuable legal case information without relying on expensive professional legal services. Our previous work CaseLink demonstrated the effectiveness of using case to case graph structures to improve retrieval accuracy. However, its high computational cost during inference on large-scale legal databases limits its practical use in real-world settings. The main inefficiency comes from constructing test time graphs and computing pairwise term frequency similarities of cases. This process has O(n^2) complexity for n legal cases, making the runtime prohibitive as the number of candidates grows. For example, the retrieval time for one query on a database (COLIEE2022) with 1,563 candidate cases is more than 500 milliseconds, while the runtime would increase drastically to more than 3,500 seconds for a database (LeCaRDv2) with 55,192 candidate cases. To further enhance the retrieval performance while achieving a significant speed-up, in this extension paper, Cassette framework is proposed with a distillation strategy involving ranking objective and eigen-matching objective for an effective transfer of knowledge from a powerful and well-trained heavy teacher retriever to a lightweight and efficient hybrid student dual encoder. Specifically, the student query encoder is implemented as a multilayer perceptron model designed for fast online processing, whereas the student candidate encoder adopts a GNN architecture, suitable for an offline manner within the case database. Extensive experiments are conducted on three benchmark datasets, and the results verify the effectiveness of the ranking distillation while achieving high efficiency. The code has been released on https://github.com/yanran-tang/Cassette/.

Related papers

• LACONIC: Dense-Level Effectiveness for Scalable Sparse Retrieval via a Two-Phase Training Curriculum [73.82125917416067]
  LACONIC is a family of learned sparse retrievers based on the Llama-3 architecture.<n>The 8B variant achieves a state-of-the-art 60.2 nDCG on the MTEB Retrieval benchmark, ranking 15th on the leaderboard.
  arXiv  Detail & Related papers  (2026-01-04T22:42:20Z)
• LegalSearchLM: Rethinking Legal Case Retrieval as Legal Elements Generation [9.894351313663874]
  Legal Case Retrieval (LCR) is a fundamental task for legal professionals.<n>Existing studies on LCR face two major limitations.<n>First, they are evaluated on relatively small-scale retrieval corpora.<n>Second, their reliance on embedding-based or lexical matching methods often results in limited representations and legally irrelevant matches.
  arXiv  Detail & Related papers  (2025-05-28T09:02:41Z)
• UQLegalAI@COLIEE2025: Advancing Legal Case Retrieval with Large Language Models and Graph Neural Networks [26.294747463024017]
  Legal case retrieval plays a pivotal role in the legal domain by facilitating the efficient identification of relevant cases.<n>The Competition on Legal Information Extraction and Entailment (COLIEE) is held annually, offering updated benchmark datasets for evaluation.<n>This paper presents a detailed description of CaseLink, the method employed by UQLegalAI, the second highest team in Task 1 of COLIEE 2025.
  arXiv  Detail & Related papers  (2025-05-27T05:32:50Z)
• Enhancing Legal Case Retrieval via Scaling High-quality Synthetic Query-Candidate Pairs [67.54302101989542]
  Legal case retrieval aims to provide similar cases as references for a given fact description. Existing works mainly focus on case-to-case retrieval using lengthy queries. Data scale is insufficient to satisfy the training requirements of existing data-hungry neural models.
  arXiv  Detail & Related papers  (2024-10-09T06:26:39Z)
• SAILER: Structure-aware Pre-trained Language Model for Legal Case Retrieval [75.05173891207214]
  Legal case retrieval plays a core role in the intelligent legal system. Most existing language models have difficulty understanding the long-distance dependencies between different structures. We propose a new Structure-Aware pre-traIned language model for LEgal case Retrieval.
  arXiv  Detail & Related papers  (2023-04-22T10:47:01Z)
• Injecting Domain Adaptation with Learning-to-hash for Effective and Efficient Zero-shot Dense Retrieval [49.98615945702959]
  We evaluate LTH and vector compression techniques for improving the downstream zero-shot retrieval accuracy of the TAS-B dense retriever. Our results demonstrate that, unlike prior work, LTH strategies when applied naively can underperform the zero-shot TAS-B dense retriever on average by up to 14% nDCG@10.
  arXiv  Detail & Related papers  (2022-05-23T17:53:44Z)
• DnS: Distill-and-Select for Efficient and Accurate Video Indexing and Retrieval [23.42790810694723]
  We propose a Knowledge Distillation framework, which we call Distill-and-Select (DnS) We train several students with different architectures and arrive at different trade-offs of performance and efficiency. Importantly, the proposed scheme allows Knowledge Distillation in large, unlabelled datasets -- this leads to good students.
  arXiv  Detail & Related papers  (2021-06-24T18:34:24Z)
• When Does Pretraining Help? Assessing Self-Supervised Learning for Law and the CaseHOLD Dataset [2.0924876102146714]
  We present a new dataset comprised of over 53,000+ multiple choice questions to identify the relevant holding of a cited case. We show that domain pretraining may be warranted when the task exhibits sufficient similarity to the pretraining corpus. Our findings inform when researchers should engage resource-intensive pretraining and show that Transformer-based architectures, too, learn embeddings suggestive of distinct legal language.
  arXiv  Detail & Related papers  (2021-04-18T00:57:16Z)
• Progressively Pretrained Dense Corpus Index for Open-Domain Question Answering [87.32442219333046]
  We propose a simple and resource-efficient method to pretrain the paragraph encoder. Our method outperforms an existing dense retrieval method that uses 7 times more computational resources for pretraining.
  arXiv  Detail & Related papers  (2020-04-30T18:09:50Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.