Embracing Diversity: A Multi-Perspective Approach with Soft Labels
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866929738082156544 |
|---|---|
| author | Muscato, Benedetta Bushipaka, Praveen Gezici, Gizem Passaro, Lucia Giannotti, Fosca Cucinotta, Tommaso |
| author_facet | Muscato, Benedetta Bushipaka, Praveen Gezici, Gizem Passaro, Lucia Giannotti, Fosca Cucinotta, Tommaso |
| contents | Prior studies show that adopting the annotation diversity shaped by different backgrounds and life experiences and incorporating them into the model learning, i.e. multi-perspective approach, contribute to the development of more responsible models. Thus, in this paper we propose a new framework for designing and further evaluating perspective-aware models on stance detection task,in which multiple annotators assign stances based on a controversial topic. We also share a new dataset established through obtaining both human and LLM annotations. Results show that the multi-perspective approach yields better classification performance (higher F1-scores), outperforming the traditional approaches that use a single ground-truth, while displaying lower model confidence scores, probably due to the high level of subjectivity of the stance detection task. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_00489 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Embracing Diversity: A Multi-Perspective Approach with Soft Labels Muscato, Benedetta Bushipaka, Praveen Gezici, Gizem Passaro, Lucia Giannotti, Fosca Cucinotta, Tommaso Computation and Language Artificial Intelligence Human-Computer Interaction Prior studies show that adopting the annotation diversity shaped by different backgrounds and life experiences and incorporating them into the model learning, i.e. multi-perspective approach, contribute to the development of more responsible models. Thus, in this paper we propose a new framework for designing and further evaluating perspective-aware models on stance detection task,in which multiple annotators assign stances based on a controversial topic. We also share a new dataset established through obtaining both human and LLM annotations. Results show that the multi-perspective approach yields better classification performance (higher F1-scores), outperforming the traditional approaches that use a single ground-truth, while displaying lower model confidence scores, probably due to the high level of subjectivity of the stance detection task. |
| title | Embracing Diversity: A Multi-Perspective Approach with Soft Labels |
| topic | Computation and Language Artificial Intelligence Human-Computer Interaction |
| url | https://arxiv.org/abs/2503.00489 |