Does Vision Accelerate Hierarchical Generalization in Neural Language Learners?
This addresses the data-efficiency gap between humans and language models in language acquisition, but the findings are incremental as they highlight limitations and suggest additional biases are needed.
The study investigated whether visual information accelerates hierarchical syntactic generalization in neural language models, finding that clear alignments between linguistic and visual components help, but unclear alignments do not, with results based on experiments using artificial and naturalistic images.
Neural language models (LMs) are arguably less data-efficient than humans from a language acquisition perspective. One fundamental question is why this human-LM gap arises. This study explores the advantage of grounded language acquisition, specifically the impact of visual information -- which humans can usually rely on but LMs largely do not have access to during language acquisition -- on syntactic generalization in LMs. Our experiments, following the poverty of stimulus paradigm under two scenarios (using artificial vs. naturalistic images), demonstrate that if the alignments between the linguistic and visual components are clear in the input, access to vision data does help with the syntactic generalization of LMs, but if not, visual input does not help. This highlights the need for additional biases or signals, such as mutual gaze, to enhance cross-modal alignment and enable efficient syntactic generalization in multimodal LMs.