SF2T: Self-supervised Fragment Finetuning of Video-LLMs for Fine-Grained Understanding

Re-assign community