ThunderSyncRL: Lossless Acceleration of Agentic Reinforcement Learning
Paper • 2610.05935 • Published • 1
Shaobai Jiang
shaobaij
AI & ML interests
None yet
Recent Activity
upvoted a paper 2 days ago
ThunderSyncRL: Lossless Acceleration of Agentic Reinforcement Learning upvoted a paper 2 days ago
Recurrent Harness for Agentic Continual Learning upvoted a paper 2 days ago
REMORY: Learning Residual Memory for Context CompactionOrganizations
None yet
upvoted 4
papers 2 days ago
Recurrent Harness for Agentic Continual Learning
Paper • 2610.12668 • Published • 27
REMORY: Learning Residual Memory for Context Compaction
Paper • 2610.11287 • Published • 171
TRACE: Rollout-Guided Quantization-Aware Training for FP4 Reinforcement Learning of MoE Language Models
Paper • 2610.07767 • Published • 92
upvoted 10
papers 3 days ago
When Do We Need On-Policy Distillation? Distilling on Offline Student Rollouts Is Often Better
Paper • 2610.11291 • Published • 28
Priming: Hybrid State Space Models From Pre-trained Transformers
Paper • 2605.08301 • Published • 1
ReplaySSM: Cache SSM Inputs, Not State
Paper • 2610.13778 • Published • 1
LoGRA: Scaling LLM Reinforcement Learning with Low-Rank Gradient Sketches
Paper • 2610.06647 • Published • 153
Fork-and-Flush: Escaping Idea Basins in Autoresearch Agents
Paper • 2610.07447 • Published • 2
GitSwarm: Decentralized Compounding Inference
Paper • 2610.04862 • Published • 2
QwenGyre: An Elastic Reinforcement Learning Framework for Training xLong-Horizon Agents
Paper • 2609.33848 • Published • 51
Harness^2: Recursive Agent Harnessing for an Open World
Paper • 2610.14224 • Published • 119
TheoPhysEval: Assessing AI agents at autonomous theoretical physics problem solving
Paper • 2610.16733 • Published • 1
Next-Concept Modeling: Fast, Parallel Coarse-to-Fine Text Generation
Paper • 2610.14162 • Published • 13
upvoted 6
papers 4 days ago
Learning to Discover Interesting Mathematics
Paper • 2609.28603 • Published • 16
Block Sparse Attention with Log-Linear Complexity
Paper • 2609.31093 • Published • 33
Your Transformer Can Hold Two Thoughts at Once: Evidence of Linear Superposition in LLMs
Paper • 2609.29845 • Published • 106
LastOPD: Taming Collapse in Latent On-Policy Distillation
Paper • 2609.28845 • Published • 31
Verifiable Hidden Dynamics Play: Generating Agentic RL Environments from Solved Mechanisms
Paper • 2609.27321 • Published • 23
Agent-Editing World Model: Rethinking World Modeling for LLM Agents
Paper • 2609.28416 • Published • 45