VideoSearch-R1: Iterative Video Retrieval and Reasoning via Soft Query Refinement
Paper • 2607.00446 • Published • 21
How to use VideoSearchR1/charades-stage2 with Transformers:
# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM
processor = AutoProcessor.from_pretrained("VideoSearchR1/charades-stage2")
model = AutoModelForMultimodalLM.from_pretrained("VideoSearchR1/charades-stage2")This is the Stage 2 VideoSearch-R1 checkpoint trained for Charades-STA, presented in the paper VideoSearch-R1: Iterative Video Retrieval and Reasoning via Soft Query Refinement.
Use with the VideoSearch-R1 codebase:
bash scripts/data_construct/download_preextracted_data.bash charades
EVAL_GPUS=0 bash scripts/inference/inference.bash charades --checkpoint VideoSearchR1/charades-stage2
Base model
Qwen/Qwen3-VL-4B-Instruct