analiza aiqbench
This study challenges the assumption that high accuracy on Chinese riddles proves genuine reasoning in large language models by isolating data contamination effects. The findings reveal a significant gap between model performance on memorized versus novel items, suggesting current frontier models often rely on pattern matching rather than true understanding of linguistic nuance. Consequently, claims about AI creativity and logical deduction must be re-evaluated with rigorous controls against training set leakage.
Reasoning or Memorization: Can LLMs Understand and Generate Chinese Xiehouyu Riddles?
arXiv:2607.23440v1 Announce Type: cross Abstract: In this paper, we push the boundary of LLM reasoning by testing them in a Chinese language game, xiehouyu, with novel xiehouyu created by linguists that had not existed before to avoid data contamination. We use multiple-choice questions (MCQ), free-form explanation generation, and new xiehouyu creation to evaluate LLMs' ability to understand and create xiehouyu. In MCQ, we use the delta of accuracy ($\Delta_{acc}$) between existing but low-frequency xiehouyu and novel ones as an index for memorization. $\Delta_{acc}$ for native speakers is very low, suggesting similar processing mechanisms. However, we found that frontier Chinese models have on average a $\Delta_{acc}$ of 23.6\%, while English-centric models tested have a mean $\Delta_{acc}$ of 5.1\%, suggesting that frontier Chinese models are likely trained with much larger Chinese data, thus memorizing more low-frequency xiehouyu. For novel xiehouyu, Gemini 3.1 Pro demonstrated remarkable ability with acc 92.6, which is 24\% higher than human accuracy. In xiehouyu creation, those created by LLMs receive much worse ratings than those by humans. These results suggest that claims about the reasoning abilities of LLMs may need careful re-examination considering the data contamination issue, and that LLMs' creativity in language-related tasks may still be behind human experts, at least in Chinese xiehouyu.
Title:Reasoning or Memorization: Can LLMs Understand and Generate Chinese Xiehouyu Riddles?
Abstract:In this paper, we push the boundary of LLM reasoning by testing them in a Chinese language game, xiehouyu, with novel xiehouyu created by linguists that had not existed before to avoid data contamination. We use multiple-choice questions (MCQ), free-form explanation generation, and new xiehouyu creation to evaluate LLMs' ability to understand and create xiehouyu. In MCQ, we use the delta of accuracy ($\Delta_{acc}$) between existing but low-frequency xiehouyu and novel ones as an index for memorization. $\Delta_{acc}$ for native speakers is very low, suggesting similar processing mechanisms. However, we found that frontier Chinese models have on average a $\Delta_{acc}$ of 23.6\%, while English-centric models tested have a mean $\Delta_{acc}$ of 5.1\%, suggesting that frontier Chinese models are likely trained with much larger Chinese data, thus memorizing more low-frequency xiehouyu. For novel xiehouyu, Gemini 3.1 Pro demonstrated remarkable ability with acc 92.6, which is 24\% higher than human accuracy. In xiehouyu creation, those created by LLMs receive much worse ratings than those by humans. These results suggest that claims about the reasoning abilities of LLMs may need careful re-examination considering the data contamination issue, and that LLMs' creativity in language-related tasks may still be behind human experts, at least in Chinese xiehouyu.
Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
arXiv:2607.23440 [cs.CL]
(or arXiv:2607.23440v1 [cs.CL] for this version)
https://doi.org/10.48550/arXiv.2607.23440 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 311 cuvinte
