Senqiao Yang

Senqiao's activity

• All
• Models
• Datasets
• Spaces
• Buckets
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

upvoted a paper 7 days ago

Mage-VL-Cowork: A Full Training Recipe to Learn Hybrid Computer Use From Evolved Harnesses

Paper • 2610.40747 • Published 14 days ago • 22

upvoted a paper 27 days ago

CarryOn: A Dual-Loop Harness for Personalized and Reliable Robot Assistance

Paper • 2610.21453 • Published 28 days ago • 12

upvoted 8 papers about 1 month ago

Native Action-Prior Learning from Videos for World Action Models

Paper • 2610.03391 • Published Oct 2 • 56

On-Policy Parameter Update Direction Underlies Generalization in LLM Post-Training

Paper • 2609.36659 • Published Sep 29 • 15

Scaling Trajectories for Complex Tasks through Recursive Self-Rewrite

Paper • 2610.02826 • Published Oct 2 • 38

FrameMorrow: Future-guided Frame Selection with Prospective Tokens for Long-Horizon Video Generation

Paper • 2609.38839 • Published Sep 30 • 50

MotorMind: Scaffolding General Vision Language Models for Zero-Shot Robot Manipulation

Paper • 2609.38078 • Published Sep 29 • 38

Does Learning Protein Folding Generalize to Broader Reasoning?

Paper • 2609.38879 • Published Sep 30 • 39

Octrees as an Explicit 3D Language

Paper • 2610.02388 • Published Oct 1 • 9

World Action Modeling with Progressive Visual Planning

Paper • 2610.02508 • Published Oct 1 • 85

upvoted 2 papers 2 months ago

ZimaBlue: Evolving Generalizable World Action Models through Scalable Video Pre-training

Paper • 2609.00188 • Published Aug 31 • 50

Beyond Data Scaling: Representation-Centric Continued Pre-training for Vision-Language-Action Models

Paper • 2608.27550 • Published Aug 27 • 70

upvoted a collection 2 months ago

VLAct

Collection

Beyond Data Scaling: Representation-Centric Continued Pre-training for Vision-Language-Action Models • 12 items • Updated Aug 31 • 3

upvoted a paper 3 months ago

StreamOPD: A Post-Training Recipe with Spatio-Temporal Cue Gating for Streaming Video Understanding

Paper • 2608.16320 • Published Aug 17 • 11

upvoted a collection 3 months ago

Mage

Collection

A family of lightweight multimodal models, including understanding and generation. • 8 items • Updated Jul 26 • 35

upvoted a paper 3 months ago

StreamArena: Toward Continuous, Interactive, and Long-Horizon Agentic Streaming Video Understanding

Paper • 2608.05703 • Published Aug 6 • 19

upvoted 3 papers 4 months ago

Mage-VL: An Efficient Codec-Native Streaming Multimodal Foundation Model

Paper • 2607.24904 • Published Jul 27 • 38

SciForma: Structure-Faithful Generation of Scientific Diagrams

Paper • 2607.18091 • Published Jul 20 • 25

Mage-Flow: An Efficient Native-Resolution Foundation Model for Image Generation and Editing

Paper • 2607.19064 • Published Jul 21 • 76

upvoted a paper 6 months ago

ParaVT: Taming the Tool Prior Paradox for Parallel Tool Use in Agentic Video Reinforcement Learning

Paper • 2605.20342 • Published May 19 • 31