CLaC at DISRPT 2025: Hierarchical Adapters for Cross-Framework Multi-lingual Discourse Relation Classification
Fuente:
arXiv
Salvato in:
| Autori principali: | , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866911166109843456 |
|---|---|
| author | Turk, Nawar Comitogianni, Daniele Kosseim, Leila |
| author_facet | Turk, Nawar Comitogianni, Daniele Kosseim, Leila |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_16903 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | CLaC at DISRPT 2025: Hierarchical Adapters for Cross-Framework Multi-lingual Discourse Relation Classification Turk, Nawar Comitogianni, Daniele Kosseim, Leila Computation and Language 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. |
| title | CLaC at DISRPT 2025: Hierarchical Adapters for Cross-Framework Multi-lingual Discourse Relation Classification |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2509.16903 |