--- pretty_name: ActiveVision Results license: cc-by-4.0 language: - en task_categories: - visual-question-answering tags: - benchmark - leaderboard - model-predictions - visual-reasoning - active-vision size_categories: - 1K...` block, compared by exact match after normalizing case, whitespace, and separators (`"C, A, G"` ≡ `"CAG"`; integer answers also match numerically, `"07"` ≡ `7`). See [`scoring.py`](https://github.com/saccharomycetes/ActiveVision/blob/main/eval/lib/scoring.py). Only final answers are released — no reasoning text and no agent tool-call traces. ## Citation ```bibtex @article{zhang2026exam, title={An exam for active observers}, author={Zhang, Jiarui and Tao, Muzi and Wang, Shangshang and Liu, Ollie and Ma, Xuezhe and Neiswanger, Willie}, journal={arXiv preprint arXiv:2607.16165}, year={2026} } ```