AI RESEARCH
A Dual-Task Paradigm to Investigate Sentence Comprehension Strategies in Language Models
arXiv CS.CL
•
ArXi:2604.26351v1 Announce Type: new Language models (LMs) behave like humans when their cognitive resources are restricted, particularly in predicting sentence processing costs such as reading times. However, it remains unclear whether such constraints similarly affect sentence comprehension strategies. Besides, existing methods do not directly target the balance between memory storage and sentence processing, which is central to human working memory.