Yukang Chen
@yukangchen_
Followers
227
Following
16
Media
10
Statuses
16
Research Scientist @NVIDIA , CS PhD from CUHK | Research in LLMs/VLMs, Long-context || Like Basketball 🏀, Cooking 🍳
Joined December 2024
Video understanding isn't just recognizing —it demands reasoning across thousands of frames. Meet Long-RL🚀 Highlights:.🧠 Dataset: LongVideo-Reason — 52K QAs with reasoning. ⚡ System: MR-SP - 2.1× faster RL for long videos. 📈 Scalability: Hour-long videos (3,600 frames) RL
2
53
264
Now it's fully open-source. We'd love your feedback and support!.🎮 Demo: 📦 Code: 🤖 Model: 🎥 Video: 📄 Paper:
arxiv.org
We introduce a full-stack framework that scales up reasoning in vision-language models (VLMs) to long videos, leveraging reinforcement learning. We address the unique challenges of long video...
0
2
5
Generate videos in just a few seconds. Try Grok Imagine, free for a limited time.
419
697
3K
🚀 We open-sourced the LongVILA-R1 model, Long-RL codebase, and an online demo!. 🔍 LongVILA-R1-7B supports both multi-choices & open QA on long videos. Achieved 70.7% on VideoMME (w/ sub). ⚡ Long-RL enables RL training on hour-long videos, with 2x RL speed-up on Qwen-VL and
1
19
126
RT @ShaotengLiu: We propose GenProp✨, a generative video propagation framework, which can seamlessly propagate any first frame edit through….
0
9
0
🚀 LongVILA is open-sourced: our comprehensive solution for scaling long-context Visual-Language Models (VLMs) to tackle the challenges of long video understanding! 🎥📖. - Paper: - Code: - Models: 🔍 What
3
19
67