Catalan258/VST-Training-Data

VST Training Data Video Streaming Thinking: VideoLLMs Can Watch and Think Simultaneously 📄 Paper | 🌐 Project Page | 💻 Code | 🤗 Models This dataset contains the full training data used for Video Streaming Thinking (VST), including both supervised fine-tuning (SFT) and reinforcement learning (RL) stages. Dataset Structure Subset Description vst_sft_data SFT data including video-text pairs from multiple sources vst_rl_data RL data for reinforcement… See the full description on the dataset page: https://huggingface.co/datasets/Catalan258/VST-Training-Data.

<iframe src="https://silytics.si/embed/item/55ce24d3-a25b-45da-b8ac-a41ba4bf0ab1" 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 Information may be incomplete or out of date.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.