William Li PRO

williamium's activity

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

upvoted a paper 23 days ago

Deep Search in Real-World Web Environments

Paper • 2610.30181 • Published 27 days ago • 108

upvoted a paper 24 days ago

ALPS: Visual State Encoders for World Models

Paper • 2610.30120 • Published 25 days ago • 23

upvoted 2 papers 25 days ago

EMBER: Predicting TCR--pMHC Binding Without a Docked Complex

Paper • 2610.27334 • Published 27 days ago • 6

Attention Sinks and Outliers in Attention Residuals

Paper • 2605.17887 • Published May 18 • 5

upvoted a paper 27 days ago

Learning to Think and Explore in 3D World

Paper • 2610.21783 • Published about 1 month ago • 62

upvoted a paper 30 days ago

3DCodeVerse: Exploring 3D Worlds as Code

Paper • 2610.22341 • Published about 1 month ago • 64

upvoted a paper about 1 month ago

Self-Retrospection Distillation: Turning Post-hoc Experiences into Prior Foresight

Paper • 2610.08077 • Published Oct 6 • 169

upvoted 6 papers 3 months ago

Co-RL: Unsupervised Reasoning Emerges from Diverse Cohort in Multi-agent RL

Paper • 2608.17253 • Published Aug 19 • 48

Demystifying Agent Skills: Why They Work-Until They Don't

Paper • 2608.14036 • Published Aug 14 • 56

Learn What's Left, Not What's Mastered: Saturation Aware Advantage Reweighting for Multi-Reward Policy Optimization

Paper • 2608.16072 • Published Aug 17 • 30

Self-Supervised Visual On-Policy Distillation

Paper • 2608.14144 • Published Aug 14 • 55

MatrAIx: Simulating the World with 8.3 Billion Persona Agents

Paper • 2608.04205 • Published Aug 4 • 61

On-Policy Self-Distillation without Any Supervision

Paper • 2608.06296 • Published Aug 9 • 124

upvoted 3 papers 4 months ago

Progress Reward Modeling for Robotic Learning: A Comprehensive Survey

Paper • 2607.21655 • Published Jul 22 • 79

Visual Contrastive Self-Distillation

Paper • 2607.21556 • Published Jul 23 • 48

From RGB Generation to Dense Field Readout: Pixel-Space Dense Prediction with Text-to-Image Models

Paper • 2607.06553 • Published Jul 9 • 15

upvoted a paper 5 months ago

SPACENUM: Revisiting Spatial Numerical Understanding in VLMs

Paper • 2605.23898 • Published May 22 • 7

upvoted 3 papers 7 months ago

CocoaBench: Evaluating Unified Digital Agents in the Wild

Paper • 2604.11201 • Published Apr 13 • 37

Can Vision Language Models Infer Human Gaze Direction? A Controlled Study

Paper • 2506.05412 • Published Jun 4, 2025 • 5

Think in Strokes, Not Pixels: Process-Driven Image Generation via Interleaved Reasoning

Paper • 2604.04746 • Published Apr 8 • 72