BLiMPPI: Benchmark of Linguistic Minimal Pairs for Polarity Items
cs.CL
Submitted: 2026-10-09
Updated: 2026-10-09
Code: https://github.com/arzuburcuguven/blimppi-data
License: http://creativecommons.org/licenses/by/4.0/
The gist: Negative polarity items (NPIs), such as "anymore" and "ever", are grammatical only in select environments, predominantly in the context of negation.
Terminology
Abstract
Negative polarity items (NPIs), such as "anymore" and "ever", are grammatical only in select environments, predominantly in the context of negation. Because their grammaticality depends on both syntax and semantics, they have a rich, multilingual literature in theoretical linguistics, yet work on NPIs in language models is almost entirely confined to English. We introduce BLiMPPI, a multilingual evaluation suite of over 12,000 NPI minimal pairs in 33 languages, built by a pipeline that discovers NPIs in Universal Dependencies (UD) treebanks, labels their licensing environments, and perturbs these into minimal pairs. Evaluating 12 multilingual and 2 monolingual models alongside n-gram baselines, we find that multilingual models struggle most with lower-resource languages and with pairs that are not n-gram solvable. We further examine the effects of NPI typology, language family, licensing variability, and NPI-licenser distance. Data: https://github.com/arzuburcuguven/blimppi-data
Sources
- Aya Expanse: Combining Research Breakthroughs for a New Multilingual Frontier
- Gemma 3 Technical Report
- The Llama 3 Herd of Models
- EuroLLM-9B: Technical Report
- Apertus: Democratizing Open and Compliant LLMs for Global Language Environments
- KoBALT: Korean Benchmark For Advanced Linguistic Tasks
- 2 OLMo 2 Furious
- Qwen3 Technical Report
Related papers
- Exploring Solution Divergence and Its Effect on Large Language Model Problem Solving
- Ishigaki-IDS-Bench: A Benchmark for Generating Information Delivery Specification from BIM Information Requirements
- Subliminal Steering: Stronger Encoding of Hidden Signals
- MedStruct-S: A Benchmark for Key Discovery, Key-Conditioned QA and Semi-Structured Extraction from OCR Clinical Reports
- The End of Transformers? On Challenging Attention and the Rise of Sub-Quadratic Architectures
- Untangling the Mechanisms of Misleading Context in Medical Question Answering