Communication Dans Un Congrès Année : 2026

Coupling Local Context and Global Semantic Prototypes via a Hierarchical Architecture for Rhetorical Roles Labeling

Résumé

Rhetorical Role Labeling (RRL) identifies the functional role of each sentence in a document, a key task for discourse understanding in domains such as law and medicine. While hierarchical models capture local dependencies effectively, they are limited in modeling global, corpus-level features. To address this limitation, we propose two prototype-based methods that integrate local context with global representations. Prototype-Based Regularization (PBR) learns soft prototypes through a distance-based auxiliary loss to structure the latent space, while Prototype-Conditioned Modulation (PCM) constructs corpus-level prototypes and injects them during training and inference. Given the scarcity of RRL resources, we introduce SCOTUS-LAW, the first dataset of U.S. Supreme Court opinions annotated with rhetorical roles at three levels of granularity: category, rhetorical function, and step. Experiments on legal, medical, and scientific benchmarks show consistent improvements over strong baselines, with ∼ 4 Macro-F1 gains on low-frequency roles. We further analyze the implications in the era of Large Language Models and complement our findings with expert evaluation.

Fichier principal
Vignette du fichier
2026.eacl-long.137 (3).pdf (3.95 Mo) Télécharger le fichier
Origine Fichiers éditeurs autorisés sur une archive ouverte
licence

Dates et versions

hal-05630007 , version 1 (22-05-2026)

Licence

Identifiants

  • HAL Id : hal-05630007 , version 1

Citer

Anas Belfathi, Nicolas Hernandez, Laura Monceaux, Warren Bonnard, Mary Catherine Lavissiere, et al.. Coupling Local Context and Global Semantic Prototypes via a Hierarchical Architecture for Rhetorical Roles Labeling. 19th Conference of the European Chapter of the Association for Computational Linguistics (EACL 2026), Mar 2026, Rabat, Morocco. pp.2986-3004. ⟨hal-05630007⟩
0 Consultations
0 Téléchargements

Partager

  • More