Challenge: Whether language models have inductive biases favoring typologically frequent grammatical properties over rare, implausible ones has been investigated, typically using artificial languages (ALs).
Approach: They extend their context-free AL formalization by adopting Generalized Categorial Grammar (GCG) . they also examine the generalization ability of LMs to process unseen longer test sentences .
Outcome: The proposed models better capture features of natural languages and can process unseen longer test sentences.

Similar Papers

How Abstract Is Linguistic Generalization in Large Language Models? Experiments with Argument Structure (2023.tacl-1)

Copied to clipboard

Challenge: Competent speakers of a language know how likely a word w is to appear in a specific context .
Approach: They use transformer-based large language models to generalize a novel noun argument . they show a bias to generalise based on linear order, instead of a linear order .
Outcome: The proposed models perform well in generalizing the distribution of a novel noun argument between related contexts that were seen during pre-training.
Exploring Compositional Generalization of Large Language Models (2024.naacl-srw)

Copied to clipboard

Challenge: a recent study has found that large language models can generalize compositional instructions from simple instructions to complex ones.
Approach: They study the generalization ability of large language models with respect to compositional instructions . they first construct a dataset with the help of ChatGPT guided by the self-instruct technique .
Outcome: The proposed model can generalize from simple instructions to more intricate ones, the authors show . their results show that training LLMs on higher-order compositional instructions improves performance on lower-order ones, but not on higher order ones.
Large Language Models Badly Generalize across Option Length, Problem Types, and Irrelevant Noun Replacements (2025.emnlp-main)

Copied to clipboard

Challenge: Existing benchmarks have exposed patterns and may not truly assess generalization ability of Large Language Models (LLMs).
Approach: They propose a “Generalization Stress Test” to assess Large Language Models’ generalization ability under slight and controlled perturbations, including option length, problem types, and irrelevant noun replacements.
Outcome: The proposed test shows that LLMs exhibit severe accuracy drops and unexpected biases when faced with minor but content-preserving modifications.
Language Models as an Alternative Evaluator of Word Order Hypotheses: A Case Study in Japanese (2020.acl-main)

Copied to clipboard

Challenge: a method using neural language models (LMs) for analyzing the word order of language is currently lacking.
Approach: They propose a method using neural language models to analyze the word order in Japanese . they test whether there is a parallel between LMs and human word order preference .
Outcome: The proposed method is validated by comparing it with other linguistic studies.
Methods for Estimating and Improving Robustness of Language Models (2022.naacl-srw)

Copied to clipboard

Challenge: Large language models suffer from weak generalisation ability due to shallow textual relations over full semantic complexity of the problem.
Approach: They propose to incorporate some of these measures into training objectives to enhance distributional robustness of LLMs.
Outcome: The proposed models outperform human models on complex tasks and outperformed other models on deep networks.
Can Language Models Learn Typologically Implausible Languages? (2026.tacl-1)

Copied to clipboard

Challenge: Language models provide a naturalistic framework for studying artificial language learning . authors: typological universals and tendencies are thought to be caused by a learning bias .
Approach: They propose to train LMs on highly naturalistic counterfactual versions of English and Japanese . they show that LM learn subtly implausible languages more slowly .
Outcome: The proposed language models learn subtly implausible languages more slowly compared to human models . the findings suggest that LMs exhibit typologically aligned learning preferences .
Do Language Models Perform Generalizable Commonsense Inference? (2021.findings-acl)

Copied to clipboard

Challenge: Recent work has applied pretrained language models to populate commonsense knowledge graphs (CKGs) but there is a lack of understanding on their generalization to multiple CKGs, unseen relations, and novel entities.
Approach: They analyze the ability of pretrained language models to perform generalizable commonsense inference in terms of knowledge capacity, transferability and induction.
Outcome: The proposed models can adapt to different schemas defined by multiple CKGs but fail to generalize to new relations.
Does Syntax Need to Grow on Trees? Sources of Hierarchical Inductive Bias in Sequence-to-Sequence Networks (2020.tacl-1)

Copied to clipboard

Challenge: Inductive biases can arise from any aspect of the model architecture, study finds . we investigate which architectural factors affect how models generalize .
Approach: They investigate which architectural factors affect generalization behavior of neural network models . they use English question formation and English tense reinflection as test cases .
Outcome: The findings suggest that human-like generalization requires architectural syntactic structure.
Logic Haystacks: Probing LLMs’ Long-Context Logical Reasoning (Without Easily Identifiable Unrelated Padding) (2026.eacl-short)

Copied to clipboard

Challenge: Recent large language models claim long context windows, but evaluations often involve simple retrieval tasks or synthetic tasks padded with irrelevant text.
Approach: They use grammars to generate simplified English with logical representations to create long input text while controlling its semantics.
Outcome: The proposed model performs better with realistic distractors than with standard models.
Anything Goes? A Crosslinguistic Study of (Im)possible Language Learning in LMs (2025.acl-long)

Copied to clipboard

Challenge: LMs are highly flexible learners, capable of acquiring linguistic patterns beyond those learnable by humans.
Approach: They train LMs to model impossible and typologically unattested languages . they find that the model does not achieve perfect separation between attested and unattest languages - suggesting some human-like inductive biases .
Outcome: The proposed model can largely distinguish attested from impossible languages, but does not achieve perfect separation between them and their impossible counterparts.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations