Vision-and-Language or Vision-for-Language? On Cross-Modal Influence in Multimodal Transformers
Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review
Standard
Vision-and-Language or Vision-for-Language? On Cross-Modal Influence in Multimodal Transformers. / Frank, Stella; Bugliarello, Emanuele; Elliott, Desmond.
Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing. Association for Computational Linguistics (ACL), 2021. p. 9847-9857.Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review
Harvard
APA
Vancouver
Author
Bibtex
}
RIS
TY - GEN
T1 - Vision-and-Language or Vision-for-Language? On Cross-Modal Influence in Multimodal Transformers
AU - Frank, Stella
AU - Bugliarello, Emanuele
AU - Elliott, Desmond
PY - 2021
Y1 - 2021
N2 - Pretrained vision-and-language BERTs aim to learn representations that combine information from both modalities. We propose a diagnostic method based on cross-modal input ablation to assess the extent to which these models actually integrate cross-modal information. This method involves ablating inputs from one modality, either entirely or selectively based on cross-modal grounding alignments, and evaluating the model prediction performance on the other modality. Model performance is measured by modality-specific tasks that mirror the model pretraining objectives (e.g. masked language modelling for text). Models that have learned to construct cross-modal representations using both modalities are expected to perform worse when inputs are missing from a modality. We find that recently proposed models have much greater relative difficulty predicting text when visual information is ablated, compared to predicting visual object categories when text is ablated, indicating that these models are not symmetrically cross-modal.
AB - Pretrained vision-and-language BERTs aim to learn representations that combine information from both modalities. We propose a diagnostic method based on cross-modal input ablation to assess the extent to which these models actually integrate cross-modal information. This method involves ablating inputs from one modality, either entirely or selectively based on cross-modal grounding alignments, and evaluating the model prediction performance on the other modality. Model performance is measured by modality-specific tasks that mirror the model pretraining objectives (e.g. masked language modelling for text). Models that have learned to construct cross-modal representations using both modalities are expected to perform worse when inputs are missing from a modality. We find that recently proposed models have much greater relative difficulty predicting text when visual information is ablated, compared to predicting visual object categories when text is ablated, indicating that these models are not symmetrically cross-modal.
U2 - 10.18653/v1/2021.emnlp-main.775
DO - 10.18653/v1/2021.emnlp-main.775
M3 - Article in proceedings
SP - 9847
EP - 9857
BT - Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing
PB - Association for Computational Linguistics (ACL)
T2 - 2021 Conference on Empirical Methods in Natural Language Processing
Y2 - 7 November 2021 through 11 November 2021
ER -
ID: 301491346