m-Just/VisCoT_VStar_Collage

This is part of the training data for vSearcher introduced in "InSight-o3: Empowering Multimodal Foundation Models with Generalized Visual Search". The data comprise collages made from a subset of images from VisualCoT and the training data of V*. Each entry of this dataset contains a collage (with a randomly placed "core" image within it) and a QA for the core image. The other images are filler images sampled from the same image pool as the core images. Every image (both core and filler) is… See the full description on the dataset page: https://huggingface.co/datasets/m-Just/VisCoT_VStar_Coll

<iframe src="https://silytics.si/embed/item/301a2863-463e-44c5-82a6-81736c7dcca5" width="420" height="190" style="border:0" loading="lazy" title="Silytics entry"></iframe>

Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 18 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.