CREATE benchmark still unsaturated: GPT-6.1-Sol hits 23.8 on creative utility
gregd_nlp · x · 2026-10-24
Updated results for CREATE, the associative-creativity benchmark proposed in March, confirm it isn't saturated: GPT-6.1-Sol scores 23.8 on creative utility, up from 18.5 for GPT-5.5 and 11.2 for GPT-4.1 — suggesting frontier models haven't hit a ceiling on creative utility yet. The authors call for more clean creativity benchmarks like this one.
More from Research
- EMNLP 2026 keynotes: Pascale Fung on world models, Neubig on open research — mohitban47 · 2026-10-24
- HarmAmp benchmark shows LLMs amplify harm over multi-turn chats; TrajSafe monitor intervenes early — alan_ritter · 2026-10-24
- 30-month study of 26,811 students: AI homework use raised scores 18% but exam scores fell 20% — lmoroney · 2026-10-24
- ~6K-param router-logit detector cuts VLM safety errors by up to 92% — shangbinfeng · 2026-10-24
- Benign image + benign question can be harmful: prompting, SFT, steering all fall short — shangbinfeng · 2026-10-24
- Fireworks explains how it made RL practical on Kimi K3 at 2.8T parameters — sophiamyang · 2026-10-24