Zero-shot Action Localization via the Confidence of Large Vision-Language Models

Re-assign community