Thunder-KoNUBench: A Corpus-Aligned Benchmark for Korean Negation Understanding
Not on the map. Why?
Abstract (arXiv)
Although negation is known to challenge large language models (LLMs), benchmarks for evaluating negation understanding-especially in Korean-are scarce. We conduct a corpus-based analysis of Korean negation and show that LLM performance degrades under negation. We then introduce Thunder-KoNUBench, a sentence-level negation understanding benchmark that reflects the empirical distribution of Korean negation phenomena. Evaluating 47 LLMs on Thunder-KoNUBench, we analyze the effects of model size and instruction tuning, and perform error analysis to better understand model behavior. We further show that fine-tuning on Thunder-KoNUBench improves negation understanding and broader contextual comprehension in Korean.
Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 26 of its 45 references that are in the map. First posted on arXiv 8 Jan 2026.
Builds on
26 of 45 references.
References · 26
- 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,084
- 2016Overcoming catastrophic forgetting in neural networkscited by 4,718
- 2018Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challengecited by 4,710
- 2019What BERT is not: Lessons from a new suite of psycholinguistic diagnostics for language modelscited by 317
- 2019Negated and Misprimed Probes for Pretrained Language Models: Birds Can Talk, But Cannot Flycited by 240
- 2019Emerging Cross-lingual Structure in Pretrained Language Modelscited by 180
- 2024KMMLU: Measuring Massive Multitask Language Understanding in Koreancited by 91
- 2021Understanding by Understanding Not: Modeling Negation in Language Modelscited by 73
- 2023Language models are not naysayers: An analysis of language models on negation benchmarkscited by 73
- 2025Vision-Language Models Do Not Understand Negationcited by 69
- 2022CONDAQA: A Contrastive Reading Comprehension Dataset for Reasoning about Negationcited by 38
- 2022KOBEST: Korean Balanced Evaluation of Significant Taskscited by 37
14 more
- 2018Do Language Models Understand Anything? On the Ability of LSTMs to Understand Negative Polarity Itemscited by 36
- 2025Know "No" Better: A Data-Driven Approach for Enhancing Negation Awareness in CLIPcited by 26
- 2025EXAONE 4.0: Unified Large Language Models Integrating Non-reasoning and Reasoning Modescited by 26
- 2022Not another Negation Benchmark: The NaN-NLI Test Suite for Sub-clausal Negationcited by 24
- 2025EXAONE Deep: Reasoning Enhanced Language Modelscited by 20
- 2020It's not a Non-Issue: Negation as a Source of Error in Machine Translationcited by 19
- 2025Kanana: Compute-efficient Bilingual Language Modelscited by 19
- 2023This is not a Dataset: A Large Negation Benchmark to Challenge Large Language Modelscited by 18
- 2019Improving Sentiment Analysis with Multi-task Learning of Negationcited by 8
- 2025Negation: A Pink Elephant in the Large Language Models' Room?cited by 7
- 2021Revisiting Negation in Neural Machine Translationcited by 6
- 2022A Question-Answer Driven Approach to Reveal Affirmative Interpretations from Verbal Negationscited by 5
- 2025Thunder-NUBench: A Benchmark for LLMs' Sentence-Level Negation Understandingcited by 5
- 2024The Self-Contained Negation Test Setcited by 3
Built on by
Not cited yet.
Similar, not cited
Close in wording, no citation either way
- 2023HAE-RAE Bench: Evaluation of Korean Knowledge in Language Models
- 2025Ko-MuSR: A Multistep Soft Reasoning Benchmark for LLMs Capable of Understanding Korean
- 2025KORMo: Korean Open Reasoning Model for Everyone
- 2023ScoNe: Benchmarking Negation Reasoning in Language Models With Fine-Tuning and In-Context Learning
- 2025KoSimpleQA: A Korean Factuality Benchmark with an Analysis of Reasoning LLMs
- 2026Optimizing Korean-Centric LLMs via Token Pruning
- 2024Pragmatic Competence Evaluation of Large Language Models for the Korean Language
- 2024KULTURE Bench: A Benchmark for Assessing Language Model in Korean Cultural Context