Cross-Modal Retrieval Augmentation for Multi-Modal Classification
Shir Gur, Natalia Neverova, Chris Stauffer, Ser-Nam Lim, Douwe Kiela, Austin Reiter
Abstract
Recent advances in using retrieval components over external knowledge sources have shown impressive results for a variety of downstream tasks in natural language processing. Here, we explore the use of unstructured external knowledge sources of images and their corresponding captions for improving visual question answering (VQA). First, we train a novel alignment model for embedding images and captions in the same space, which achieves substantial improvement in performance on image-caption retrieval w.r.t. similar methods. Second, we show that retrieval-augmented multi-modal transformers using the trained alignment model improve results on VQA over strong baselines. We further conduct extensive experiments to establish the promise of this approach, and examine novel applications for inference time such as hot-swapping indices.- Anthology ID:
- 2021.findings-emnlp.11
- Volume:
- Findings of the Association for Computational Linguistics: EMNLP 2021
- Month:
- November
- Year:
- 2021
- Address:
- Punta Cana, Dominican Republic
- Editors:
- Marie-Francine Moens, Xuanjing Huang, Lucia Specia, Scott Wen-tau Yih
- Venue:
- Findings
- SIG:
- SIGDAT
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 111–123
- Language:
- URL:
- https://aclanthology.org/2021.findings-emnlp.11
- DOI:
- 10.18653/v1/2021.findings-emnlp.11
- Cite (ACL):
- Shir Gur, Natalia Neverova, Chris Stauffer, Ser-Nam Lim, Douwe Kiela, and Austin Reiter. 2021. Cross-Modal Retrieval Augmentation for Multi-Modal Classification. In Findings of the Association for Computational Linguistics: EMNLP 2021, pages 111–123, Punta Cana, Dominican Republic. Association for Computational Linguistics.
- Cite (Informal):
- Cross-Modal Retrieval Augmentation for Multi-Modal Classification (Gur et al., Findings 2021)
- PDF:
- https://preview.aclanthology.org/ingest-acl-2023-videos/2021.findings-emnlp.11.pdf
- Data
- Flickr30k