KoNeoBench - Contextual Identification without Particles: leaderboard
KoNeoBench evaluates Korean neologism understanding: contextual identification, component source-word recovery, semantic and specialist-domain categorization, and definition generation. Table 6 identifies the instruction-tuned checkpoints behind shortened result labels. This series reports KoNeoBench - Contextual Identification without Particles from 2609.19916v1, S4.T1: Accuracy (%). Higher is better.
Metric: Accuracy (%). Source: arxiv.org. Saturation forecast: Around May 2027. 9 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.4 | 66.16 | #85 |
| 2 | GPT-4.1 | 54.62 | #260 |
| 3 | Qwen 3.5 9B | 47.4 | #395 |
| 4 | EXAONE 3.5 7.8B Instruct | 47.09 | #862 |
| 5 | Solar Pro 3 | 44.34 | #576 |
| 6 | Qwen 2.5 7B Instruct | 40.47 | #866 |
| 7 | Llama 3.1 8B Instruct | 34.47 | #1021 |
No result here: #3 Claude Opus 5.5, #5 GPT-6 Astra, #6 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=koneobench-contextual-identification-without-particles · How It Works · Data refreshed daily, snapshot 2026-11-07.