Yilun Zhao PRO

yilunzhao's activity

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

upvoted a paper 4 days ago

Watch Before You Answer: Learning from Visually Grounded Post-Training

Paper • 2604.05117 • Published 9 days ago • 35

upvoted a paper 5 days ago

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

Paper • 2604.04746 • Published 7 days ago • 70

upvoted a paper 6 days ago

Learning to Retrieve from Agent Trajectories

Paper • 2604.04949 • Published 16 days ago • 69

upvoted a paper 27 days ago

AI Can Learn Scientific Taste

Paper • 2603.14473 • Published about 1 month ago • 422

upvoted 4 papers about 1 month ago

LLM2Vec-Gen: Generative Embeddings from Large Language Models

Paper • 2603.10913 • Published Mar 11 • 44

OpenClaw-RL: Train Any Agent Simply by Talking

Paper • 2603.10165 • Published Mar 10 • 151

RbtAct: Rebuttal as Supervision for Actionable Review Feedback Generation

Paper • 2603.09723 • Published Mar 10 • 7

Reasoning Models Struggle to Control their Chains of Thought

Paper • 2603.05706 • Published Mar 5 • 37

upvoted 7 papers about 2 months ago

LongCLI-Bench: A Preliminary Benchmark and Study for Long-horizon Agentic Programming in Command-Line Interfaces

Paper • 2602.14337 • Published Feb 15 • 15

Learning from Trials and Errors: Reflective Test-Time Planning for Embodied LLMs

Paper • 2602.21198 • Published Feb 24 • 4

On Data Engineering for Scaling LLM Terminal Capabilities

Paper • 2602.21193 • Published Feb 24 • 102

MedXIAOHE: A Comprehensive Recipe for Building Medical MLLMs

Paper • 2602.12705 • Published Feb 13 • 68

Less is Enough: Synthesizing Diverse Data in Feature Space of LLMs

Paper • 2602.10388 • Published Feb 11 • 244

SkillsBench: Benchmarking How Well Agent Skills Work Across Diverse Tasks

Paper • 2602.12670 • Published Feb 13 • 59

Data Darwinism Part I: Unlocking the Value of Scientific Data for Pre-training

Paper • 2602.07824 • Published Feb 8 • 18

upvoted 5 papers 2 months ago

How2Everything: Mining the Web for How-To Procedures to Evaluate and Improve LLMs

Paper • 2602.08808 • Published Feb 9 • 9

ANCHOR: Branch-Point Data Generation for GUI Agents

Paper • 2602.07153 • Published Feb 6 • 5

SAGE: Benchmarking and Improving Retrieval for Deep Research Agents

Paper • 2602.05975 • Published Feb 5 • 12

SWE-World: Building Software Engineering Agents in Docker-Free Environments

Paper • 2602.03419 • Published Feb 3 • 41

PaperSearchQA: Learning to Search and Reason over Scientific Papers with RLVR

Paper • 2601.18207 • Published Jan 26 • 19