Home ➤ Publications ➤ arulanandam2026section

Section-Weighted Hybrid Approach for Legal Case Retrieval

Rajith Arulanandam, Nisansa De Silva
2026 8th International Conference on Natural Language Processing (ICNLP)

Finding truly analogous precedents requires capturing legal reasoning beyond surface word overlap. We present a two-stage, section-aware framework for legal case retrieval that first segments raw judgments into facts, issues, decision, and reasoning using a deterministic large language model (LLM) offline. In Stage 1, we combine parallel lexical (BM25) and semantic (dense ANN) whole-document searches via Reciprocal Rank Fusion (RRF) to form a high-recall candidate pool. In Stage 2, we perform fine-grained, like-for-like comparisons (e.g., query reasoning vs. candidate reasoning). To address the scale mismatch between unbounded lexical scores and cosine similarities, we apply query-wise Z-score normalization before aggregating signals with learned section weights. For the top results, the system returns the relevant section text with a concise, grounded rationale and party-stance labels. We evaluate on a jurisdiction-scale benchmark, demonstrating consistent gains over strong lexical and neural baselines while maintaining high candidate coverage.

@inproceedings{arulanandam2026section,
  title={{Section-Weighted Hybrid Approach for Legal Case Retrieval}},
  author={Arulanandam, Rajith and De Silva, Nisansa},
  booktitle={2026 8th International Conference on Natural Language Processing (ICNLP)},
  pages={567--571},
  year={2026},
  abstract={Finding truly analogous precedents requires capturing legal reasoning beyond surface word overlap. We present a two-stage, section-aware framework for legal case retrieval that first segments raw judgments into facts, issues, decision, and reasoning using a deterministic large language model (LLM) offline. In Stage 1, we combine parallel lexical (BM25) and semantic (dense ANN) whole-document searches via Reciprocal Rank Fusion (RRF) to form a high-recall candidate pool. In Stage 2, we perform fine-grained, like-for-like comparisons (e.g., query reasoning vs. candidate reasoning). To address the scale mismatch between unbounded lexical scores and cosine similarities, we apply query-wise Z-score normalization before aggregating signals with learned section weights. For the top results, the system returns the relevant section text with a concise, grounded rationale and party-stance labels. We evaluate on a jurisdiction-scale benchmark, demonstrating consistent gains over strong lexical and neural baselines while maintaining high candidate coverage.},
  doi={10.1109/ICNLP69856.2026.11527762},
  organization={IEEE},
  misc={https://arxiv.org/pdf/2606.03138,NLP:ML:LAW,https://goo.gl/iY6aTr}
}

Related Projects

SigmaLaw

Active

PI: Nisansa de Silva

This project delves into the synergy between Natural Language Processing (NLP) and the legal domain, presenting novel solutions to enduring challenges.