# XCOMPS: A Multilingual Benchmark of Conceptual Minimal Pairs

arXiv: https://arxiv.org/abs/2502.19737 · HTML: https://mlsanity.com/paper/2502.19737/

## Key facts

- Cited by 12 papers in the map (data through 14 Oct 2026).
- Builds on 31 of its 55 references that are in the map.
- First posted on arXiv 27 Feb 2025.

## Builds on (most cited first, 30 of 31)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://mlsanity.com/paper/1910.10683/) (2019)
- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter](https://mlsanity.com/paper/1910.01108/) (2019)
- [Unsupervised Cross-lingual Representation Learning at Scale](https://mlsanity.com/paper/1911.02116/) (2019)
- [Finetuned Language Models Are Zero-Shot Learners](https://mlsanity.com/paper/2109.01652/) (2021)
- [TinyBERT: Distilling BERT for Natural Language Understanding](https://mlsanity.com/paper/1909.10351/) (2019)
- [Cross-lingual Language Model Pretraining](https://mlsanity.com/paper/1901.07291/) (2019)
- [XNLI: Evaluating Cross-lingual Sentence Representations](https://mlsanity.com/paper/1809.05053/) (2018)
- [XTREME: A Massively Multilingual Multi-task Benchmark for Evaluating Cross-lingual Generalization](https://mlsanity.com/paper/2003.11080/) (2020)
- [BLiMP: The Benchmark of Linguistic Minimal Pairs for English](https://mlsanity.com/paper/1912.00582/) (2019)
- [Measuring and Improving Consistency in Pretrained Language Models](https://mlsanity.com/paper/2102.01017/) (2021)
- [Assessing the Ability of LSTMs to Learn Syntax-Sensitive Dependencies](https://mlsanity.com/paper/1611.01368/) (2016)
- [MLQA: Evaluating Cross-lingual Extractive Question Answering](https://mlsanity.com/paper/1910.07475/) (2019)
- [PAWS-X: A Cross-lingual Adversarial Dataset for Paraphrase Identification](https://mlsanity.com/paper/1908.11828/) (2019)
- [Targeted Syntactic Evaluation of Language Models](https://mlsanity.com/paper/1808.09031/) (2018)
- [A Systematic Assessment of Syntactic Generalization in Neural Language Models](https://mlsanity.com/paper/2005.03692/) (2020)
- [WikiLingua: A New Benchmark Dataset for Cross-Lingual Abstractive Summarization](https://mlsanity.com/paper/2010.03093/) (2020)
- [Neural Language Models as Psycholinguistic Subjects: Representations of Syntactic State](https://mlsanity.com/paper/1903.03260/) (2019)
- [Multilingual LAMA: Investigating Knowledge in Multilingual Pretrained Language Models](https://mlsanity.com/paper/2102.00894/) (2021)
- [MLSUM: The Multilingual Summarization Corpus](https://mlsanity.com/paper/2004.14900/) (2020)
- [Cross-Lingual Consistency of Factual Knowledge in Multilingual Language Models](https://mlsanity.com/paper/2310.10378/) (2023)
- [Cross-Linguistic Syntactic Evaluation of Word Prediction Models](https://mlsanity.com/paper/2005.00187/) (2020)
- [CLiMP: A Benchmark for Chinese Language Model Evaluation](https://mlsanity.com/paper/2101.11131/) (2021)
- [Event knowledge in large language models: the gap between the impossible and the unlikely](https://mlsanity.com/paper/2212.01488/) (2022)
- [Decoding Probing: Revealing Internal Linguistic Structures in Neural Language Models using Minimal Pairs](https://mlsanity.com/paper/2403.17299/) (2024)
- [RuBLiMP: Russian Benchmark of Linguistic Minimal Pairs](https://mlsanity.com/paper/2406.19232/) (2024)
- [DaLAJ - a dataset for linguistic acceptability judgments for Swedish: Format, baseline, sharing](https://mlsanity.com/paper/2105.06681/) (2021)
- [Is Incoherence Surprising? Targeted Evaluation of Coherence Prediction from Language Models](https://mlsanity.com/paper/2105.03495/) (2021)

## Built on by (most cited first, 12 of 12)

- [Do Construction Distributions Shape Formal Language Learning In German BabyLMs?](https://mlsanity.com/paper/2503.11593/) (2025)
- [BabyBabelLM: A Multilingual Benchmark of Developmentally Plausible Training Data](https://mlsanity.com/paper/2510.10159/) (2025)
- [Large Language Model probabilities cannot distinguish between possible and impossible language](https://mlsanity.com/paper/2509.15114/) (2025)
- [From Chains to DAGs: Probing the Graph Structure of Reasoning in LLMs](https://mlsanity.com/paper/2601.17593/) (2026)
- [Structural priors for data-efficient language learning](https://mlsanity.com/paper/2609.11505/) (2026)
- [The First ChineseBabyLM Challenge: training data-efficient and cognitively plausible language models for Chinese](https://mlsanity.com/paper/2607.10745/) (2026)
- [Grammatical "grandmother neurons" are rare in LLMs](https://mlsanity.com/paper/2609.29328/) (2026)
- [Feeding BabyLMs Macaroni: Code-Switching Curricula Cause Cross-Lingual Convergence](https://mlsanity.com/paper/2609.30535/) (2026)
- [Ontology-Grounded, Reasoner-Verified Benchmarks for Evaluating LLM Reasoning in Scientific AI](https://mlsanity.com/paper/2610.00682/) (2026)
- [Introducing Code-Switched Contexts to Cognitively-Inspired Bilingual Model Training](https://mlsanity.com/paper/2610.06161/) (2026)
- [Language Exposure Matters: Distribution and Order in Multilingual BabyLM Pre-training](https://mlsanity.com/paper/2610.20102/) (2026)
- [Scaling Multilingual Mixture-of-Experts: How Language Coverage Changes Capacity and Compute Allocation](https://mlsanity.com/paper/2610.20247/) (2026)

## Abstract (arXiv)

We introduce XCOMPS in this work, a multilingual conceptual minimal pair dataset covering 17 languages. Using this dataset, we evaluate LLMs' multilingual conceptual understanding through metalinguistic prompting, direct probability measurement, and neurolinguistic probing. By comparing base, instruction-tuned, and knowledge-distilled models, we find that: 1) LLMs exhibit weaker conceptual understanding for low-resource languages, and accuracy varies across languages despite being tested on the same concept sets. 2) LLMs excel at distinguishing concept-property pairs that are visibly different but exhibit a marked performance drop when negative pairs share subtle semantic similarities. 3) Instruction tuning improves performance in concept understanding but does not enhance internal competence; knowledge distillation can enhance internal competence in conceptual understanding for low-resource languages with limited gains in explicit task performance. 4) More morphologically complex languages yield lower concept understanding scores and require deeper layers for conceptual reasoning.