New Benchmark Tests Vision-Language Models on Atari Gameplay Reasoning
A new arXiv paper, AtariBench, evaluates how well vision-language models handle spatiotemporal reasoning during Atari gameplay, moving beyond static visual understanding tests.
Read the full story at arXiv cs.CV ↗
Timeline · 1 report
- 2026-11-10 05:00 · arXiv cs.CV
AtariBench: Evaluating Spatiotemporal Reasoning of Vision-Language Models in Atari Gameplay