Z-LaVI: Zero-Shot Language Solver Fueled by Visual Imagination
Yue Yang, Wenlin Yao, Hongming Zhang, Xiaoyang Wang, Dong Yu, Jianshu Chen
Abstract
Large-scale pretrained language models have made significant advances in solving downstream language understanding tasks. However, they generally suffer from reporting bias, the phenomenon describing the lack of explicit commonsense knowledge in written text, e.g., ”an orange is orange”. To overcome this limitation, we develop a novel approach, Z-LaVI, to endow language models with visual imagination capabilities. Specifically, we leverage two complementary types of ”imaginations”: (i) recalling existing images through retrieval and (ii) synthesizing nonexistent images via text-to-image generation. Jointly exploiting the language inputs and the imagination, a pretrained vision-language model (e.g., CLIP) eventually composes a zero-shot solution to the original language tasks. Notably, fueling language models with imagination can effectively leverage visual knowledge to solve plain language tasks. In consequence, Z-LaVI consistently improves the zero-shot performance of existing language models across a diverse set of language tasks.- Anthology ID:
- 2022.emnlp-main.78
- Volume:
- Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing
- Month:
- December
- Year:
- 2022
- Address:
- Abu Dhabi, United Arab Emirates
- Editors:
- Yoav Goldberg, Zornitsa Kozareva, Yue Zhang
- Venue:
- EMNLP
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 1186–1203
- Language:
- URL:
- https://aclanthology.org/2022.emnlp-main.78
- DOI:
- 10.18653/v1/2022.emnlp-main.78
- Cite (ACL):
- Yue Yang, Wenlin Yao, Hongming Zhang, Xiaoyang Wang, Dong Yu, and Jianshu Chen. 2022. Z-LaVI: Zero-Shot Language Solver Fueled by Visual Imagination. In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, pages 1186–1203, Abu Dhabi, United Arab Emirates. Association for Computational Linguistics.
- Cite (Informal):
- Z-LaVI: Zero-Shot Language Solver Fueled by Visual Imagination (Yang et al., EMNLP 2022)
- PDF:
- https://preview.aclanthology.org/landing_page/2022.emnlp-main.78.pdf