AI RESEARCH

Filling in the Mechanisms: How do LMs Learn Filler-Gap Dependencies under Developmental Constraints?

arXiv CS.CL

ArXi:2604.14459v1 Announce Type: new For humans, filler-gap dependencies require a shared representation across different syntactic constructions. Although causal analyses suggest this may also be true for LLMs (Bogurae, 2025), it is still unclear if such a representation also exists for language models trained on developmentally feasible quantities of data.