Distilling Knowledge from Text-to-Image Generative Models Improves Visio-Linguistic Reasoning in CLIP

Samyadeep Basu, Shell Xu Hu, Maziar Sanjabi, Daniela Massiceti, Soheil Feizi


Abstract
Image-text contrastive models like CLIP have wide applications in zero-shot classification, image-text retrieval, and transfer learning. However, they often struggle on compositional visio-linguistic tasks (e.g., attribute-binding or object-relationships) where their performance is no better than random chance. To address this, we introduce SDS-CLIP, a lightweight and sample-efficient distillation method to enhance CLIP’s compositional visio-linguistic reasoning. Our approach fine-tunes CLIP using a distillation objective borrowed from large text-to-image generative models like Stable-Diffusion, which are known for their strong visio-linguistic reasoning abilities. On the challenging Winoground benchmark, SDS-CLIP improves the visio-linguistic performance of various CLIP models by up to 7%, while on the ARO dataset, it boosts performance by up to 3%. This work underscores the potential of well-designed distillation objectives from generative models to enhance contrastive image-text models with improved visio-linguistic reasoning capabilities.
Anthology ID:
2024.emnlp-main.351
Volume:
Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing
Month:
November
Year:
2024
Address:
Miami, Florida, USA
Editors:
Yaser Al-Onaizan, Mohit Bansal, Yun-Nung Chen
Venue:
EMNLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
6105–6113
Language:
URL:
https://aclanthology.org/2024.emnlp-main.351
DOI:
Bibkey:
Cite (ACL):
Samyadeep Basu, Shell Xu Hu, Maziar Sanjabi, Daniela Massiceti, and Soheil Feizi. 2024. Distilling Knowledge from Text-to-Image Generative Models Improves Visio-Linguistic Reasoning in CLIP. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, pages 6105–6113, Miami, Florida, USA. Association for Computational Linguistics.
Cite (Informal):
Distilling Knowledge from Text-to-Image Generative Models Improves Visio-Linguistic Reasoning in CLIP (Basu et al., EMNLP 2024)
Copy Citation:
PDF:
https://aclanthology.org/2024.emnlp-main.351.pdf