ACL Findings2025
Large Language Models as Neurolinguistic Subjects: Discrepancy between Performance and Competence
Linyang He, Ercong Nie, Helmut Schmid, Hinrich Schütze, Nima Mesgarani, Jonathan Brennan
Abstract
This study investigates the linguistic understanding of Large Language Models (LLMs) regarding signifier (form) and signified (meaning) by distinguishing two LLM assessment paradigms: psycholinguistic and neurolinguistic. Traditional psycholinguistic evaluations often reflect statistical rules that may not accurately represent LLMs' true linguistic competence. We introduce a neurolinguistic approach, utilizing a novel method that combines minimal pairs and diagnostic probing to analyze activation patterns across model layers. This method allows for a detailed examination of how LLMs represent form and meaning, and whether these representations are consistent across languages. We found: (1) Psycholinguistic and neurolinguistic methods reveal that language performance and competence are distinct; (2) Direct probability measurement may not accurately assess linguistic competence; (3) Instruction tuning won't change much competence but improve performance; (4) LLMs exhibit higher competence and performance in form compared to meaning. Additionally, we introduce new conceptual minimal pair datasets for Chinese (COMPS-ZH) and German (COMPS-DE), complementing existing English datasets. 1
Ask Lune
Know this paper in minutes, not an afternoon.
Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.
2 top-tier papers have cited it since. Know what they found before your reviewers do.
Lune can draw on
the full paper · the top-tier papers that cite it · related work
Free to start. No credit card required.
Lineage
Cited by top-tier papers2
- Far from the Shallow: Brain-Predictive Reasoning Embedding through Residual DisentanglementLinyang He, Tianjun Zhong, Richard J. Antonello, Gavin Mischler et al.NeurIPS 2025 · 6 citations
- Temporal Precision Matters: Brain-Tuning Speech Language Models with Millisecond-Resolution Neural SignalsZhejun Zhang, Wenqing Zhou, Haozhe Xu, Lin Zhang et al.ACL 2026
Builds on6
- Climbing towards NLU: On Meaning, Form, and Understanding in the Age of DataEmily M. Bender, Alexander KollerACL 2020 · 914 citations
- ChatGPT Beyond English: Towards a Comprehensive Evaluation of Large Language Models in Multilingual LearningViet Dac Lai, Nghia Trung Ngo, Amir Pouran Ben Veyseh, Hieu Man et al.EMNLP Findings 2023 · 233 citations
- A Systematic Assessment of Syntactic Generalization in Neural Language ModelsJennifer Hu, Jon Gauthier, Peng Qian, Ethan Wilcox et al.ACL 2020 · 124 citations
- Prompting is not a substitute for probability measurements in large language modelsJennifer Hu, Roger LevyEMNLP 2023 · 31 citations
- Cross-Lingual Retrieval Augmented Prompt for Low-Resource LanguagesErcong Nie, Sheng Liang, Helmut Schmid, Hinrich SchützeACL Findings 2023 · 21 citations
Related papers
- CxMP: A Linguistic Minimal-Pair Benchmark for Evaluating Constructional Understanding in Language ModelsMiyu Oba, Saku SugawaraACL 2026
- From Language to Cognition: How LLMs Outgrow the Human Language NetworkBadr AlKhamissi, Greta Tuckute, Yingtian Tang, Taha Osama A Binhuraib et al.EMNLP 2025 · 1 citation
- How Hypocritical Is Your LLM judge? Listener-Speaker Asymmetries in the Pragmatic Competence of Large Language ModelsJudith Sieker, Sina ZarrießACL Findings 2026
- Disentangling Language and Culture for Evaluating Multilingual Large Language ModelsJiahao Ying, Wei Tang, Yiran Zhao, Yixin Cao et al.ACL 2025 · 7 citations
- Interpretability of Language Models via Task SpacesLucas Weber, Jaap Jumelet, Elia Bruni, Dieuwke HupkesACL 2024
