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

TL;DR

This paper introduces prototype-based methods to enhance hierarchical models for Rhetorical Role Labeling by integrating local context with global semantic prototypes, improving performance especially on low-frequency roles.

Contribution

It proposes two novel prototype-based techniques for RRL, introduces the SCOTUS-Law dataset, and demonstrates improved results across multiple domains.

Findings

4 Macro-F1 gain on low-frequency roles

Consistent improvements over strong baselines

Effective integration of local and global features

Abstract

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.