analiza aiqbench
This cs.CL study tests whether LLMs generalize a novel verb from active to passive voice the way people do. Humans extend the form-based pattern even when the verb has no plausible meaning, while the models succeed only when the verb can be mapped onto a familiar one. For AI readers, it's a useful benchmark-style result suggesting that current LLM syntactic competence may depend more on lexical distribution and meaning than on abstract structural rules.
Examining form-based linguistic productivity in people and large language models
arXiv:2610.33433v1 Announce Type: new Abstract: Linguistic productivity, as evidenced by the use of words in novel contexts, is one of the most striking facts about human language. However, how abstract the knowledge that enables such productivity is remains unclear. We compare people's and computational language models' linguistic productivity in contexts where successful generalization depends on form-based information. We focus on extending the use of a novel verb learned in the active voice to passive contexts. We find people extend active to passive uses in a way consistent with abstract generalizations concerning linguistic form, even when no plausible meaning can be assigned to the novel verb. In contrast, large language models succeeded only when a meaning like an existing verb's could be assigned to the novel verb. We conclude that computational models' syntactic knowledge remains tied to the distribution and meaning of specific lexical items in a way that human syntactic knowledge is not.
Abstract:Linguistic productivity, as evidenced by the use of words in novel contexts, is one of the most striking facts about human language. However, how abstract the knowledge that enables such productivity is remains unclear. We compare people's and computational language models' linguistic productivity in contexts where successful generalization depends on form-based information. We focus on extending the use of a novel verb learned in the active voice to passive contexts. We find people extend active to passive uses in a way consistent with abstract generalizations concerning linguistic form, even when no plausible meaning can be assigned to the novel verb. In contrast, large language models succeeded only when a meaning like an existing verb's could be assigned to the novel verb. We conclude that computational models' syntactic knowledge remains tied to the distribution and meaning of specific lexical items in a way that human syntactic knowledge is not.
Computation and Language (cs.CL)
arXiv:2610.33433 [cs.CL]
(or arXiv:2610.33433v1 [cs.CL] for this version)
https://doi.org/10.48550/arXiv.2610.33433 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 243 cuvinte
Subiecte conexe
Linguistic Productivity in Large Language Models: Models Coerce, but do not Preempt
arXiv · 3 iun. 2026
Linguistic Productivity in Large Language Models: Models Coerce, but do not Preempt
arXiv · 22 oct. 2026
How Human-Like Are Large Language Models? A Register-Aware Linguistic Evaluation Framework
arXiv · 1 sept. 2026
DecompSR: A dataset for decomposed analyses of compositional multihop spatial reasoning
arXiv · 12 iun. 2026