TurBLiMP: Turkish BLIMP

Dataset Description

TurBLiMP is the first Turkish benchmark of linguistic minimal pairs, designed to evaluate the linguistic abilities of monolingual and multilingual language models. The dataset covers 16 core grammatical phenomena in Turkish, with 1,000 minimal pairs per phenomenon.

Dataset Structure

The dataset contains minimal pairs of grammatical and ungrammatical sentences in Turkish, organized into subsets testing various linguistic phenomena. Each minimal pair tests a specific grammatical phenomenon.

Fields

• sentence_good: The grammatical sentence
• sentence_bad: The ungrammatical sentence
• linguistic_phenomenon: The linguistic phenomenon being tested
• good_cue: The cue word in the grammatical sentence
• bad_cue: The cue word in the ungrammatical sentence
• critical_region: The critical region being tested
• item_id: Unique identifier for the item

Phenomena

1. Anaphor Agreement - Reflexive pronoun agreement violations
2. Argument Structure (Transitive) - Case marking errors with transitive verbs
3. Argument Structure (Ditransitive) - Case marking errors with ditransitive verbs
4. Binding - Principle B violations in binding theory
5. Determiners - Obligatory use of the indefinite article
6. Ellipsis - Backward gapping with non-parallel word orders
7. Irregular Forms - Incorrect aorist allomorph usage
8. Island Effects - Wh-adjunct extraction from complex NPs
9. Nominalization - Incorrect nominalization suffix selection
10. NPI Licensing - Negative polarity items in non-negative contexts
11. Passives - Unlicensed use of by-phrases in impersonal passives
12. Quantifiers - Quantifier usage with bare nouns
13. Relative Clauses - Incorrect case marking in relative clauses
14. Scrambling - Illicit postverbal scrambling from embedded clauses
15. Subject Agreement - Person/number agreement violations
16. Suspended Affixation - Improper tense suffix suspension

Usage

from datasets import load_dataset

# Load specific subset
dataset = load_dataset("juletxara/turblimp", "anaphor_agreement")

# Or load all subsets
dataset = load_dataset("juletxara/turblimp")

# Example usage
for subset_name, subset_data in dataset.items():
    print(f"Subset: {subset_name}")
    for example in subset_data:
        print(f"Good: {example['sentence_good']}")
        print(f"Bad: {example['sentence_bad']}")
        print(f"Phenomenon: {example['linguistic_phenomenon']}")
        print()

Data Source

This dataset is derived from TurBLiMP: https://github.com/ezgibasar/TurBLiMP

Citation

Please cite the original TurBLiMP paper when using this dataset:

@misc{başar2025turblimpturkishbenchmarklinguistic,
  title={TurBLiMP: A Turkish Benchmark of Linguistic Minimal Pairs}, 
  author={Ezgi Başar and Francesca Padovani and Jaap Jumelet and Arianna Bisazza},
  year={2025},
  eprint={2506.13487},
  archivePrefix={arXiv},
  primaryClass={cs.CL},
  url={https://arxiv.org/abs/2506.13487}
}

License

The dataset follows the original TurBLiMP project's CC BY 4.0 licensing terms.