Approximate Attributions for Off-the-Shelf Siamese Transformers

Lucas Möller, Dmitry Nikolaev, Sebastian Padó

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Siamese encoders such as sentence transformers are among the least understood deep models. Established attribution methods cannot tackle this model class since it compares two inputs rather than processing a single one. To address this gap, we have recently proposed an attribution method specifically for Siamese encoders (Möller et al., 2023). However, it requires models to be adjusted and fine-tuned and therefore cannot be directly applied to off-the-shelf models. In this work, we reassess these restrictions and propose (i) a model with exact attribution ability that retains the original model's predictive performance and (ii) a way to compute approximate attributions for off-the-shelf models. We extensively compare approximate and exact attributions and use them to analyze the models' attendance to different linguistic aspects. We gain insights into which syntactic roles Siamese transformers attend to, confirm that they mostly ignore negation, explore how they judge semantically opposite adjectives, and find that they exhibit lexical bias.
Original languageEnglish
Title of host publicationProceedings of the 18th Conference of the European Chapter of the Association for Computational Linguistics
Publication statusAccepted/In press - 18 Jan 2024

Fingerprint

Dive into the research topics of 'Approximate Attributions for Off-the-Shelf Siamese Transformers'. Together they form a unique fingerprint.

Cite this