SVLTA: Benchmarking Vision-Language Temporal Alignment via Synthetic Video Situation

Re-assign community