Qian Wu

Fivethousand's activity

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

upvoted 2 papers 2 days ago

StreamDiT: Real-Time Streaming Text-to-Video Generation

Paper • 2507.03745 • Published 6 days ago • 23

MemOS: A Memory OS for AI System

Paper • 2507.03724 • Published 6 days ago • 101

upvoted 2 papers 3 days ago

Arch-Router: Aligning LLM Routing with Human Preferences

Paper • 2506.16655 • Published 20 days ago • 10

SPIRAL: Self-Play on Zero-Sum Games Incentivizes Reasoning via Multi-Agent Multi-Turn Reinforcement Learning

Paper • 2506.24119 • Published 10 days ago • 43

upvoted 2 papers 2 months ago

Towards Understanding Camera Motions in Any Video

Paper • 2504.15376 • Published Apr 21 • 157

Subject-driven Video Generation via Disentangled Identity and Motion

Paper • 2504.17816 • Published Apr 23 • 11

upvoted 6 papers 3 months ago

Packing Input Frame Context in Next-Frame Prediction Models for Video Generation

Paper • 2504.12626 • Published Apr 17 • 52

C3PO: Critical-Layer, Core-Expert, Collaborative Pathway Optimization for Test-Time Expert Re-Mixing

Paper • 2504.07964 • Published Apr 10 • 61

VisualCloze: A Universal Image Generation Framework via Visual In-Context Learning

Paper • 2504.07960 • Published Apr 10 • 49

Self-Steering Language Models

Paper • 2504.07081 • Published Apr 9 • 18

Hogwild! Inference: Parallel LLM Generation via Concurrent Attention

Paper • 2504.06261 • Published Apr 8 • 110

One-Minute Video Generation with Test-Time Training

Paper • 2504.05298 • Published Apr 7 • 106

upvoted 3 papers 4 months ago

Scalable Language Models with Posterior Inference of Latent Thought Vectors

Paper • 2502.01567 • Published Feb 3 • 1

Bridging Continuous and Discrete Tokens for Autoregressive Visual Generation

Paper • 2503.16430 • Published Mar 20 • 34

MPBench: A Comprehensive Multimodal Reasoning Benchmark for Process Errors Identification

Paper • 2503.12505 • Published Mar 16 • 10

upvoted 2 papers 5 months ago

MotionLab: Unified Human Motion Generation and Editing via the Motion-Condition-Motion Paradigm

Paper • 2502.02358 • Published Feb 4 • 19

Analyze Feature Flow to Enhance Interpretation and Steering in Language Models

Paper • 2502.03032 • Published Feb 5 • 61

upvoted a paper 12 months ago

NeedleBench: Can LLMs Do Retrieval and Reasoning in 1 Million Context Window?

Paper • 2407.11963 • Published Jul 16, 2024 • 45