eyehole/VisChainBench
VisChainBench VisChainBench is a benchmark for multi-turn image-centric reasoning, designed to evaluate and test visual language models (VLMs) across various tasks and scenarios. Existing benchmarks increasingly emphasise multi-image understanding, they largely focus on static, language-heavy comparisons (e.g., "Compare Image 1 and 2" or "Is Image 3 suitable for Image 2?"), neglecting the progressive, context-dependent reasoning that defines human problem-solving. VisChainBench… See the full description on the dataset page: https://huggingface.co/datasets/eyehole/VisChainBench.
| Author / organization | eyehole |
|---|---|
| License (as declared) | cc |
| Downloads | 355 (as reported by Hugging Face) |
| Likes | 3 |
| Created | 5 May 2025 |
| Last updated at source | 20 May 2025 |
Recent changes
- 16 Oct 2026 Description: VisChainBench VisChainBench is a benchmark for multi-turn image-centric reasoning, designed to evaluate and test visual language models (VLMs) across various tasks and scenarios. Existing benchmarks increasingly emphasise multi-image unders → VisChainBench VisChainBench is a benchmark for multi-turn image-centric reasoning, designed to evaluate and test visual language models (VLMs) across various tasks and scenarios. Existing benchmarks increasingly emphasise multi-image unders
Changes detected by Silytics between refreshes of the source.
Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 16 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.