Abstract
Post-hoc explanation methods are an important tool for increasing model transparency for users. Unfortunately, the currently used methods for attributing token importance often yield diverging patterns. In this work, we study potential sources of disagreement across methods from a linguistic perspective. We find that different methods systematically select different classes of words and that methods that agree most with other methods and with humans display similar linguistic preferences. Token-level differences between methods are smoothed out if we compare them on the syntactic span level. We also find higher agreement across methods by estimating the most important spans dynamically instead of relying on a fixed subset of size k. We systematically investigate the interaction between k and spans and propose an improved configuration for selecting important tokens.- Anthology ID:
- 2024.lrec-main.1397
- Volume:
- Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024)
- Month:
- May
- Year:
- 2024
- Address:
- Torino, Italia
- Editors:
- Nicoletta Calzolari, Min-Yen Kan, Veronique Hoste, Alessandro Lenci, Sakriani Sakti, Nianwen Xue
- Venues:
- LREC | COLING
- SIG:
- Publisher:
- ELRA and ICCL
- Note:
- Pages:
- 16066–16078
- Language:
- URL:
- https://preview.aclanthology.org/remove-affiliations/2024.lrec-main.1397/
- DOI:
- Cite (ACL):
- Jonathan Kamp, Lisa Beinborn, and Antske Fokkens. 2024. The Role of Syntactic Span Preferences in Post-Hoc Explanation Disagreement. In Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024), pages 16066–16078, Torino, Italia. ELRA and ICCL.
- Cite (Informal):
- The Role of Syntactic Span Preferences in Post-Hoc Explanation Disagreement (Kamp et al., LREC-COLING 2024)
- PDF:
- https://preview.aclanthology.org/remove-affiliations/2024.lrec-main.1397.pdf