Chengsong Huang

ChengsongHuang's activity

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

upvoted a paper 1 day ago

UniEvo-VL: An On-policy Self-Distillation Training Recipe for Multimodal Model Self-improvement

Paper • 2609.38721 • Published 4 days ago • 274

upvoted a paper 2 days ago

On the Off-Policy Teacher in On-Policy Distillation

Paper • 2609.38360 • Published 5 days ago • 19

upvoted 3 papers 3 days ago

TabFM: A Zero-Shot Foundation Model for Tabular Data

Paper • 2609.37959 • Published 5 days ago • 17

Adversarial Training for Pixel Diffusion

Paper • 2609.38170 • Published 5 days ago • 18

TGRL: Temperature-Grouped Reinforcement Learning for Efficient Exploration in LLMs

Paper • 2609.33589 • Published 7 days ago • 8

upvoted a paper 5 days ago

FuseReg: Regularizing Layer Fusion Mitigates the Reconstruction-Generation Gap in Representation Autoencoders

Paper • 2609.31620 • Published 9 days ago • 159

authored a paper 10 days ago

RRSI: Regularized Recursive Self-Improvement of Agent Harnesses

Paper • 2609.24972 • Published 13 days ago • 221

authored 6 papers 11 days ago

Small RL Controller, Large Language Model: RL-Guided Adaptive Sampling for Test-Time Scaling

Paper • 2606.03102 • Published Jun 2 • 14

MoCo: A One-Stop Shop for Model Collaboration Research

Paper • 2601.21257 • Published Apr 19

You Only Need Minimal RLVR Training: Extrapolating LLMs via Rank-1 Trajectories

Paper • 2605.21468 • Published May 20 • 51

Process Rewards with Learned Reliability

Paper • 2605.15529 • Published May 15 • 52

G-Zero: Self-Play for Open-Ended Generation from Zero Data

Paper • 2605.09959 • Published May 11 • 18

LLMs Improving LLMs: Agentic Discovery for Test-Time Scaling

Paper • 2605.08083 • Published May 8 • 71

upvoted a paper 11 days ago

RRSI: Regularized Recursive Self-Improvement of Agent Harnesses

Paper • 2609.24972 • Published 13 days ago • 221

upvoted a paper 12 days ago

Designer-RSI: Evolving Procedural Memory from User Traffic for Agentic Graphic Design

Paper • 2609.22086 • Published 16 days ago • 34

New activity in IntelligenceLab/LHTB-leaderboard 12 days ago

LHTB 2.0 / candidate-77: four Vertex runs, with full verifier output

#12 opened 12 days ago by

ChengsongHuang

Add Longer-Horizon Terminal Bench 2.0 / candidate-77: four Vertex AI runs

#11 opened 12 days ago by

ChengsongHuang

upvoted 2 papers 18 days ago

Dream-RSI: Recursive Self-Improvement through Evolving Worlds

Paper • 2609.14858 • Published 20 days ago • 251

Benchmark Radar: A Living Database and Search Engine for AI Benchmarks and Evaluation

Paper • 2609.11115 • Published 24 days ago • 174

upvoted a paper 22 days ago

Recursive Code World Models: Building Complex Worlds through Recursive Scene Programs

Paper • 2609.11499 • Published 24 days ago • 33