# CLaC at DISRPT 2025: Hierarchical Adapters for Cross-Framework Multi-lingual Discourse Relation Classification

arXiv: https://arxiv.org/abs/2509.16903 · HTML: https://mlsanity.com/paper/2509.16903/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 12 of its 56 references that are in the map.
- First posted on arXiv 21 Sep 2025.

## Builds on (most cited first, 12 of 12)

- [Supervised Learning of Universal Sentence Representations from Natural Language Inference Data](https://mlsanity.com/paper/1705.02364/) (2017)
- [SciDTB: Discourse Dependency TreeBank for Scientific Abstracts](https://mlsanity.com/paper/1806.03653/) (2018)
- [GENTLE: A Genre-Diverse Multilayer Challenge Set for English NLP and Linguistic Evaluation](https://mlsanity.com/paper/2306.01966/) (2023)
- [DisCoDisCo at the DISRPT2021 Shared Task: A System for Discourse Segmentation, Classification, and Connective Detection](https://mlsanity.com/paper/2109.09777/) (2021)
- [GCDT: A Chinese RST Treebank for Multigenre and Multilingual Discourse Parsing](https://mlsanity.com/paper/2210.10449/) (2022)
- [Unifying Discourse Resources with Dependency Framework](https://mlsanity.com/paper/2101.00167/) (2021)
- [Persian Rhetorical Structure Theory](https://mlsanity.com/paper/2106.13833/) (2021)
- [Zero-shot Chinese Discourse Dependency Parsing via Cross-lingual Mapping](https://mlsanity.com/paper/1911.12014/) (2019)
- [A Multi-Task and Multi-Label Classification Model for Implicit Discourse Relation Recognition](https://mlsanity.com/paper/2408.08971/) (2024)
- [Probing LLMs for Multilingual Discourse Generalization Through a Unified Label Set](https://mlsanity.com/paper/2503.10515/) (2025)
- [Constrained Multi-Layer Contrastive Learning for Implicit Discourse Relationship Recognition](https://mlsanity.com/paper/2409.13716/) (2024)
- [Multi-Lingual Implicit Discourse Relation Recognition with Multi-Label Hierarchical Learning](https://mlsanity.com/paper/2508.20712/) (2025)

## Abstract (arXiv)

We present our submission to Task 3 (Discourse Relation Classification) of the DISRPT 2025 shared task. Task 3 introduces a unified set of 17 discourse relation labels across 39 corpora in 16 languages and six discourse frameworks, posing significant multilingual and cross-formalism challenges. We first benchmark the task by fine-tuning multilingual BERT-based models (mBERT, XLM-RoBERTa-Base, and XLM-RoBERTa-Large) with two argument-ordering strategies and progressive unfreezing ratios to establish strong baselines. We then evaluate prompt-based large language models (namely Claude Opus 4.0) in zero-shot and few-shot settings to understand how LLMs respond to the newly proposed unified labels. Finally, we introduce HiDAC, a Hierarchical Dual-Adapter Contrastive learning model. Results show that while larger transformer models achieve higher accuracy, the improvements are modest, and that unfreezing the top 75% of encoder layers yields performance comparable to full fine-tuning while training far fewer parameters. Prompt-based models lag significantly behind fine-tuned transformers, and HiDAC achieves the highest overall accuracy (67.5%) while remaining more parameter-efficient than full fine-tuning.