Abstract
We argue that language-only models don’t learn the physical manifestation of language. We present an empirical investigation of visual-auditory properties of language through a series of tasks, termed H-Test.These tasks highlight a fundamental gap between human linguistic understanding and the sensory-deprived linguistic understanding of LLMs. In support of our hypothesis, 1. deliberate reasoning (Chain-of-Thought), 2. few-shot examples, or 3. stronger LLM from the same model family (LLaMA 2 13B -> LLaMA 2 70B) has no significant effect on H-Test performance. We bring in the philosophical case of Mary, who learns about the world in a sensory-deprived environment as a useful conceptual framework to understand how language-only models learn about the world (Jackson, 1986). Our experiments show that some of the strongest proprietary LLMs stay near random chance baseline accuracy of 50%, highlighting the limitations of linguistic knowledge acquired in the absence of sensory experience. Our code and data are available at <github.com/brucewlee/h-test>.- Anthology ID:
- 2024.acl-long.195
- Volume:
- Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
- Month:
- August
- Year:
- 2024
- Address:
- Bangkok, Thailand
- Editors:
- Lun-Wei Ku, Andre Martins, Vivek Srikumar
- Venue:
- ACL
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 3554–3579
- Language:
- URL:
- https://preview.aclanthology.org/icon-24-ingestion/2024.acl-long.195/
- DOI:
- 10.18653/v1/2024.acl-long.195
- Cite (ACL):
- Bruce Lee and Jaehyuk Lim. 2024. Language Models Don’t Learn the Physical Manifestation of Language. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 3554–3579, Bangkok, Thailand. Association for Computational Linguistics.
- Cite (Informal):
- Language Models Don’t Learn the Physical Manifestation of Language (Lee & Lim, ACL 2024)
- PDF:
- https://preview.aclanthology.org/icon-24-ingestion/2024.acl-long.195.pdf