🔄 In a Training Loop

Peng Wang

stillarrow

·

https://peter-peng-w.github.io/

AI & ML interests

None yet

Recent Activity

liked a model 2 days ago

amazon/chronos-2

updated a model 4 days ago

stillarrow/qwen2.5-math-7b__cr7b_scpo_w0_seed1

published a model 4 days ago

stillarrow/qwen2.5-math-7b__cr7b_scpo_w0_seed1

View all activity

Organizations

None yet

upvoted a paper 18 days ago

Self-Evolving Coding Agents via Harness-Policy Optimization

Paper • 2610.22791 • Published 22 days ago • 49

upvoted a collection about 1 month ago

Qwen3

84 items • Updated Dec 31, 2025 • 1.87k

upvoted a paper about 1 month ago

Rethinking Critic Learning in PPO: Understanding and Mitigating Value Flattening

Paper • 2609.18708 • Published Sep 16 • 56

upvoted a paper 2 months ago

Your Language Model is Its Own Critic: Reinforcement Learning with Value Estimation from Actor's Internal States

Paper • 2605.07579 • Published May 8 • 19

upvoted a paper 3 months ago

From RLVR to RLSVR: Task Transformation Induces Self-Verifiable Rewards for Open-Ended LLM Self-Improvement

Paper • 2607.23802 • Published Jul 26 • 79

upvoted 2 articles 4 months ago

Article

NEO-unify: Building Native Multimodal Unified Models End to End

sensenova

•

Mar 5

• 182

Article

Welcome Gemma 4: Frontier multimodal intelligence on device

+5

merve, pcuenq, sergiopaniego, burtenshaw, Steveeeeeeen, alvarobartt, SaylorTwift

•

Apr 2

• 929

upvoted 5 papers 6 months ago

TradingAgents: Multi-Agents LLM Financial Trading Framework

Paper • 2412.20138 • Published Dec 28, 2024 • 163

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

Paper • 2605.21468 • Published May 20 • 51

MetaAgent-X : Breaking the Ceiling of Automatic Multi-Agent Systems via End-to-End Reinforcement Learning

Paper • 2605.14212 • Published May 14 • 19

GDPO: Group reward-Decoupled Normalization Policy Optimization for Multi-reward RL Optimization

Paper • 2601.05242 • Published Jan 8 • 235

Agent World Model: Infinity Synthetic Environments for Agentic Reinforcement Learning

Paper • 2602.10090 • Published Feb 10 • 53

upvoted 3 papers 7 months ago

HY-World 2.0: A Multi-Modal World Model for Reconstructing, Generating, and Simulating 3D Worlds

Paper • 2604.14268 • Published Apr 15 • 125

Heterogeneous Agent Collaborative Reinforcement Learning

Paper • 2603.02604 • Published Mar 3 • 94

Self-Distilled RLVR

Paper • 2604.03128 • Published Apr 3 • 141

upvoted a collection 8 months ago

Qwen2.5-Coder

Code-specific model series based on Qwen2.5 • 38 items • Updated Mar 2 • 379

upvoted 2 papers 8 months ago

PaperBanana: Automating Academic Illustration for AI Scientists

Paper • 2601.23265 • Published Jan 30 • 231

MiroThinker-1.7 & H1: Towards Heavy-Duty Research Agents via Verification

Paper • 2603.15726 • Published Mar 16 • 186

upvoted 2 collections 8 months ago

NeMo Gym

Collection of RL verifiable data for NeMo Gym • 32 items • Updated 17 days ago • 64

BFS-Prover

LLM Step-Provers in Lean4 • 5 items • Updated Oct 7, 2025 • 9