# Thunder-KoNUBench: A Corpus-Aligned Benchmark for Korean Negation Understanding

arXiv: https://arxiv.org/abs/2601.04693 · HTML: https://mlsanity.com/paper/2601.04693/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 26 of its 45 references that are in the map.
- First posted on arXiv 8 Jan 2026.

## Builds on (most cited first, 26 of 26)

- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge](https://mlsanity.com/paper/1803.05457/) (2018)
- [Overcoming catastrophic forgetting in neural networks](https://mlsanity.com/paper/1612.00796/) (2016)
- [What BERT is not: Lessons from a new suite of psycholinguistic diagnostics for language models](https://mlsanity.com/paper/1907.13528/) (2019)
- [Negated and Misprimed Probes for Pretrained Language Models: Birds Can Talk, But Cannot Fly](https://mlsanity.com/paper/1911.03343/) (2019)
- [Emerging Cross-lingual Structure in Pretrained Language Models](https://mlsanity.com/paper/1911.01464/) (2019)
- [KMMLU: Measuring Massive Multitask Language Understanding in Korean](https://mlsanity.com/paper/2402.11548/) (2024)
- [Language models are not naysayers: An analysis of language models on negation benchmarks](https://mlsanity.com/paper/2306.08189/) (2023)
- [Understanding by Understanding Not: Modeling Negation in Language Models](https://mlsanity.com/paper/2105.03519/) (2021)
- [Vision-Language Models Do Not Understand Negation](https://mlsanity.com/paper/2501.09425/) (2025)
- [CONDAQA: A Contrastive Reading Comprehension Dataset for Reasoning about Negation](https://mlsanity.com/paper/2211.00295/) (2022)
- [KOBEST: Korean Balanced Evaluation of Significant Tasks](https://mlsanity.com/paper/2204.04541/) (2022)
- [Do Language Models Understand Anything? On the Ability of LSTMs to Understand Negative Polarity Items](https://mlsanity.com/paper/1808.10627/) (2018)
- [EXAONE 4.0: Unified Large Language Models Integrating Non-reasoning and Reasoning Modes](https://mlsanity.com/paper/2507.11407/) (2025)
- [Know "No" Better: A Data-Driven Approach for Enhancing Negation Awareness in CLIP](https://mlsanity.com/paper/2501.10913/) (2025)
- [Not another Negation Benchmark: The NaN-NLI Test Suite for Sub-clausal Negation](https://mlsanity.com/paper/2210.03256/) (2022)
- [EXAONE Deep: Reasoning Enhanced Language Models](https://mlsanity.com/paper/2503.12524/) (2025)
- [Kanana: Compute-efficient Bilingual Language Models](https://mlsanity.com/paper/2502.18934/) (2025)
- [It's not a Non-Issue: Negation as a Source of Error in Machine Translation](https://mlsanity.com/paper/2010.05432/) (2020)
- [This is not a Dataset: A Large Negation Benchmark to Challenge Large Language Models](https://mlsanity.com/paper/2310.15941/) (2023)
- [Improving Sentiment Analysis with Multi-task Learning of Negation](https://mlsanity.com/paper/1906.07610/) (2019)
- [Negation: A Pink Elephant in the Large Language Models' Room?](https://mlsanity.com/paper/2503.22395/) (2025)
- [Revisiting Negation in Neural Machine Translation](https://mlsanity.com/paper/2107.12203/) (2021)
- [A Question-Answer Driven Approach to Reveal Affirmative Interpretations from Verbal Negations](https://mlsanity.com/paper/2205.11467/) (2022)
- [Thunder-NUBench: A Benchmark for LLMs' Sentence-Level Negation Understanding](https://mlsanity.com/paper/2506.14397/) (2025)
- [The Self-Contained Negation Test Set](https://mlsanity.com/paper/2408.11469/) (2024)

## 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.