arXiv：cs.CL· SungHo Kim, Eojin Jeon, SangKeun Lee·· 9 小时前
KoAGE：大语言模型能否识别、学习并创造韩语新造词？
KoAGE: Do Large Language Models Know, Learn, and Coin Korean Neologisms?
AI 导读
研究者提出 NeoTrace 框架与 KoAGE 基准，用 4,271 条韩语新造词评测模型对新词的识别、学习与创造能力。在 21 个模型的测试中，模型仅能定义 13-30% 的截止年份词汇，却可从单次用例推断词义；多智能体造词更偏好缩略形式，且依赖交互设定。
正文
Abstract:Korean neologisms arise through diverse mechanisms, a demanding test of language models beyond lexical memorization. To enable systematic evaluation, we introduce NeoTrace, an automated, rerunnable framework that collects and dates neologisms across genres. Its 4,271 entries support KoAGE, a cutoff-aware benchmark with three acquisition and two generation tasks---the latter, to our knowledge, the first to compare model coinages with attested neologisms under their original materials and community context. Across 21 models, lexical recall diverges from contextual inference: in a five-model temporal analysis, models define only 13-30% of cutoff-year words, yet can infer meaning from one usage. However, structural understanding remains limited: even correct definitions barely improve formation analysis of pre-cutoff neologisms. Given word-building materials, single models often follow human formation mechanisms, whereas multi-agent coinage favors abbreviation and depends on interaction framing. KoAGE thus separates what models know, learn from use, and coin.
| Subjects: | Computation and Language (cs.CL); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.19399 [cs.CL] |
| (or arXiv:2610.19399v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2610.19399 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: SungHo Kim [view email]
[v1]
Wed, 14 Oct 2026 01:49:19 UTC (2,009 KB)
来源：arXiv：cs.CL · arxiv.org