RoTA-Bench: Evaluating 3D Spatial Reasoning through Interactive Projection Alignment
Paper • 2610.17665 • Published • 207
Dataset Viewer
The dataset viewer is not available because its heuristics could not detect any supported data files. You can try uploading some data files, or configuring the data files location manually.
RoTA-Bench Evaluate
Hugging Face Evaluate integration for RoTA-Bench. The module downloads the public dataset and runs the released evaluator.
Usage
pip install evaluate huggingface_hub
import evaluate
metric = evaluate.load("QingchuanMa/rota-bench-eval")
result = metric.compute(backend="simulated", max_tasks=1)
print(result)
For an API-backed evaluation, set OPENAI_API_KEY in the environment and pass backend="gemini_api", base_url, and api_model. The API key is never stored in this repository.
The default simulated call runs one task. Increase max_tasks for a larger run. Formal comparison settings and the fixed dataset are documented at QingchuanMa/RoTA-Bench.
Paper
RoTA-Bench paper: https://huggingface.co/papers/2610.17665
- Downloads last month
- 48