koiwave/hle-short
HLE Short Dataset 100 shortest questions without images from the HLE (Humanity's Last Exam) dataset. A carefully curated subset designed for efficient evaluation and benchmarking of AI models with a focus on concise, text-only questions. 📊 Dataset Overview Source: cais/hle test split Size: 100 samples (carefully selected from 2,500 original samples) Format: Arrow format (preserves all data including binary image fields) Modalities: Text-only (all images filtered… See the full description on the dataset page: https://huggingface.co/datasets/koiwave/hle-short.
| Author / organization | koiwave |
|---|---|
| License (as declared) | mit |
| Downloads | 565 (as reported by Hugging Face) |
| Likes | 1 |
| Created | 15 August 2025 |
| Last updated at source | 15 August 2025 |
Popularity, last 3 days
565 now (-13 since 2026-10-16). Counts as reported by the source, snapshotted daily by Silytics.
Recent changes
- 17 Oct 2026 Description: HLE Short Dataset 100 shortest questions without images from the HLE (Humanity's Last Exam) dataset. A carefully curated subset designed for efficient evaluation and benchmarking of AI models with a focus on concise, text-only questions. 📊 → HLE Short Dataset 100 shortest questions without images from the HLE (Humanity's Last Exam) dataset. A carefully curated subset designed for efficient evaluation and benchmarking of AI models with a focus on concise, text-only questions. 📊
Changes detected by Silytics between refreshes of the source.
More from koiwave
- koiwave/hle-liteDataset
Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 18 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.