Retrieval-Based Multi-Label Legal Annotation: Extensible, Data-Efficient and Hallucination-Free

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhang, Li, Savelka, Jaromir, Ashley, Kevin
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914571847991296
author Zhang, Li
Savelka, Jaromir
Ashley, Kevin
author_facet Zhang, Li
Savelka, Jaromir
Ashley, Kevin
contents Multi-label legal annotation requires assigning multiple labels from large, evolving taxonomies to long, fact-intensive documents, often under limited supervision. Parametric encoders typically require task-specific training and retraining when the label set changes, while prompting generative large language models becomes costly and degrades as the label space grows. We cast legal annotation as retrieval: we embed documents and label descriptions with a frozen retrieval model and predict labels via k-nearest neighbors in the embedding space, enabling updates by re-embedding and re-indexing rather than gradient-based backpropagation. Across three legal datasets (ECtHR-A, ECtHR-B, and Eurlex with 100 labels), retrieval achieves competitive accuracy and strong data efficiency; on Eurlex, Qwen-8B retrieval improves Macro-F1 from 40.41 (GPT-5.2, zero-shot) to 49.12 while reducing estimated compute by 20-30 times compared to fine-tuning. With only (N=100) training samples, retrieval nearly doubles Micro-F1 over hierarchical Legal-BERT on ECtHR-A (48.29 vs. 27.87). We also quantify a reliability failure mode of generative inference: GPT-5.2 hallucinates labels outside the provided taxonomy in 0.12-0.9% of test samples under deterministic decoding. In contrast, retrieval strictly respects defined label sets, eliminating hallucination by design. These results suggest retrieval-model-based annotators are a practical, deployable alternative for high-cardinality and rapidly changing legal label spaces.
format Preprint
id arxiv_https___arxiv_org_abs_2605_16767
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Retrieval-Based Multi-Label Legal Annotation: Extensible, Data-Efficient and Hallucination-Free
Zhang, Li
Savelka, Jaromir
Ashley, Kevin
Computation and Language
68T50
I.2.7; I.2.4
Multi-label legal annotation requires assigning multiple labels from large, evolving taxonomies to long, fact-intensive documents, often under limited supervision. Parametric encoders typically require task-specific training and retraining when the label set changes, while prompting generative large language models becomes costly and degrades as the label space grows. We cast legal annotation as retrieval: we embed documents and label descriptions with a frozen retrieval model and predict labels via k-nearest neighbors in the embedding space, enabling updates by re-embedding and re-indexing rather than gradient-based backpropagation. Across three legal datasets (ECtHR-A, ECtHR-B, and Eurlex with 100 labels), retrieval achieves competitive accuracy and strong data efficiency; on Eurlex, Qwen-8B retrieval improves Macro-F1 from 40.41 (GPT-5.2, zero-shot) to 49.12 while reducing estimated compute by 20-30 times compared to fine-tuning. With only (N=100) training samples, retrieval nearly doubles Micro-F1 over hierarchical Legal-BERT on ECtHR-A (48.29 vs. 27.87). We also quantify a reliability failure mode of generative inference: GPT-5.2 hallucinates labels outside the provided taxonomy in 0.12-0.9% of test samples under deterministic decoding. In contrast, retrieval strictly respects defined label sets, eliminating hallucination by design. These results suggest retrieval-model-based annotators are a practical, deployable alternative for high-cardinality and rapidly changing legal label spaces.
title Retrieval-Based Multi-Label Legal Annotation: Extensible, Data-Efficient and Hallucination-Free
topic Computation and Language
68T50
I.2.7; I.2.4
url https://arxiv.org/abs/2605.16767