Xirui Li PRO

AIcell's activity

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

upvoted a paper 11 days ago

Representation Redistribution in Sparse Autoencoders

Paper • 2610.21240 • Published 25 days ago • 9

upvoted 2 papers 28 days ago

Towards Physically Grounded Articulated 3D Assets: A Multi-Agent Pipeline and Benchmark

Paper • 2610.13947 • Published Oct 9 • 14

Embodied Turing Machines: Stateful Code for Robot Recursive Self-Improvement

Paper • 2610.12369 • Published Oct 8 • 25

upvoted a paper 29 days ago

RobotWorld: Benchmarking Multimodal Agents for Robot Use Across Diverse Tasks and Embodiments

Paper • 2610.10409 • Published Oct 7 • 35

upvoted 4 papers about 1 month ago

4DCodeBench: Benchmarking Agents on Inverse Graphics of Dynamic Scenes

Paper • 2610.03715 • Published Oct 2 • 19

Decoding Looped Transformers Better for (Almost) Free

Paper • 2610.02185 • Published Oct 1 • 24

Sharpening Tax in Post-Training

Paper • 2610.01509 • Published Oct 1 • 77

4Director: Controlling Video World Models with Rigid 3D Geometry

Paper • 2610.02160 • Published Oct 1 • 18

upvoted a collection about 1 month ago

Guava

Collection

An effective and universal harness for embodied manipulation • 3 items • Updated Sep 30 • 1

upvoted a paper about 1 month ago

LEGO-Anything: Coding Agents for 3D Scene Reconstruction

Paper • 2609.36380 • Published Sep 28 • 66

upvoted 3 papers about 2 months ago

CodeMidas: Scaling Agentic Coding RL Environments from Code Itself

Paper • 2609.22068 • Published Sep 18 • 116

RRSI: Regularized Recursive Self-Improvement of Agent Harnesses

Paper • 2609.24972 • Published Sep 21 • 164

Transferring the Intelligence of VLMs to Robotic Control

Paper • 2609.22966 • Published Sep 19 • 60

upvoted 4 papers 2 months ago

Code World Model: Coding Agent as World Brain

Paper • 2608.25927 • Published Aug 26 • 43

What Makes Good Agentic Data? An ACE Lens on Data Generation for LLM Agents

Paper • 2608.27260 • Published Aug 27 • 72

Code as Worlds: Agentic Discovery of Executable World Representations for Physical Reasoning

Paper • 2608.27549 • Published Aug 27 • 48

Game2World Engine: Unlocking In-the-Wild Gameplay Videos for World Model Training

Paper • 2608.24680 • Published Aug 25 • 11

upvoted 3 papers 3 months ago

Zetta ζ: An Efficient Closed-Loop Embodied Harness for Self-Evolving Physical Intelligence

Paper • 2608.16590 • Published Aug 17 • 141

aDSL: Agentic 3D Creation via Joint Agent-Program Design

Paper • 2608.17975 • Published Aug 18 • 7

TRACE-Bench: Decomposing and Diagnosing Multi-Reference Image Generation

Paper • 2608.16765 • Published Aug 17 • 13