[2604.04825] Plausibility as Commonsense Reasoning: Humans Succeed, Large Language Models Do not

Computer Science > Computation and Language arXiv:2604.04825 (cs) [Submitted on 6 Apr 2026] Title:Plausibility as Commonsense Reasoning: Humans Succeed, Large Language Models Do not Authors:Sercan Karakaş View a PDF of the paper titled Plausibility as Commonsense Reasoning: Humans Succeed, Large Language Models Do not, by Sercan Karaka\c{s} View PDF HTML (experimental) Abstract:Large language models achieve strong performance on many language tasks, yet it remains unclear whether they integrate world knowledge with syntactic structure in a human-like, structure-sensitive way during ambiguity resolution. We test this question in Turkish prenominal relative-clause attachment ambiguities, where the same surface string permits high attachment (HA) or low attachment (LA). We construct ambiguous items that keep the syntactic configuration fixed and ensure both parses remain pragmatically possible, while graded event plausibility selectively favors High Attachment vs.\ Low Attachment. The contrasts are validated with independent norming ratings. In a speeded forced-choice comprehension experiment, humans show a large, correctly directed plausibility effect. We then evaluate Turkish and multilingual LLMs in a parallel preference-based setup that compares matched HA/LA continuations via mean per-token log-probability. Across models, plausibility-driven shifts are weak, unstable, or reversed. The results suggest that, in the tested models, plausibility information does not guide att...