CUHK
We track 8 benchmarks that CUHK made or helped make.
1 in use 8 with public code 5 with a leaderboard here
Benchmarks
| Name | Kind | Released | Best score | Status |
|---|---|---|---|---|
| VisFactor Models take 20 vision-centric subtests adapted from a human cognitive psychology test battery, covering perception primitives and visual memory. Scored by accuracy. | Spatial reasoning and counting | Feb 2025 | 55.9% | In use |
| Math-VR Models answer 5,000 English and Chinese math questions whose solutions need visual reasoning, such as drawing auxiliary lines or plotting functions. Answers are checked against ground truth. | Visual math and diagrams | Oct 2025 | | System cards only |
| VideoGAIA Agents inspect selected video segments, search the web, and read pages before answering questions. Model judges check answers, and the harness reports accuracy by category and difficulty. | Video understanding | Aug 2026 | | Not in cards |
| BlenderWorld Visual agents turn a text, image or video brief into an editable Blender scene with rendered media. Hidden binary rubrics and eight visual quality dimensions score each project. | CAD models | Oct 2026 | 79.9 | Not in cards |
| MathIF Reasoning models solve 420 math problems from GSM8K, MATH-500, Minerva, Olympiad and AIME while obeying added verifiable instructions. Code checks the instructions and the answer. | Following precise instructions | May 2025 | | Not in cards |
| PhoneHarness Bench Agents complete Android workflows through graphical controls, device commands, and structured tools. Checks inspect files, system settings, application state, and unintended side effects. | Phones | Jun 2026 | | Not in cards |
| CPRet Models retrieve contest solutions, duplicate problems, or full problems from simplified descriptions. Four retrieval tasks score the ranking of relevant results with NDCG@10. | Competitive programming | May 2025 | Not in cards | |
| Embodied Agent Arena Vision-language agents solve geometry, spatial reasoning, affordance, planning, and manipulation tasks. The 1,000-case evaluation combines 32 established sources with GeoProbe. | Embodied reasoning | Oct 2026 | Not in cards |