LISA VLM finetuned using AVS-Bench dataset
Multimodal Test-time Adaptation Framework for Visual Search