VideoSearch-R1: Iterative Video Retrieval and Reasoning via Soft Query Refinement
Paper • 2607.00446 • Published • 16
VideoSearch-R1 DiDeMo Stage 2
This is the Stage 2 VideoSearch-R1 checkpoint trained for DiDeMo, presented in the paper VideoSearch-R1: Iterative Video Retrieval and Reasoning via Soft Query Refinement.
- Project Page: mlvlab.github.io/VideoSearch-R1
- Repository: GitHub - mlvlab/VideoSearch-R1
Usage
Use with the VideoSearch-R1 codebase:
bash scripts/data_construct/download_preextracted_data.bash didemo
EVAL_GPUS=0 bash scripts/inference/inference.bash didemo --checkpoint VideoSearchR1/didemo-stage2
Citation
@inproceedings{lee2026videosearchr1,
title = {VideoSearch-R1: Iterative Video Retrieval and Reasoning via Soft Query Refinement},
author = {Lee, Seohyun and Choi, Seoung and Ko, Dohwan and Kim, Jongha and Kim, Hyunwoo J.},
booktitle = {European Conference on Computer Vision (ECCV)},
year = {2026}
}
- Downloads last month
- 75
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support
Model tree for VideoSearchR1/didemo-stage2
Base model
Qwen/Qwen3-VL-4B-Instruct
# Load model directly from transformers import AutoProcessor, AutoModelForMultimodalLM processor = AutoProcessor.from_pretrained("VideoSearchR1/didemo-stage2") model = AutoModelForMultimodalLM.from_pretrained("VideoSearchR1/didemo-stage2")