TRACE: A Unified Rollout Budget Allocation Framework for Efficient Agentic Reinforcement Learning
Paper • 2606.11119 • Published • 18
Heming Zou
gfyddha
·
AI & ML interests
None yet
Recent Activity
submitted a paper 30 days ago
TRACE: A Unified Rollout Budget Allocation Framework for Efficient Agentic Reinforcement Learning authored a paper 30 days ago
FlyLoRA: Boosting Task Decoupling and Parameter Efficiency via Implicit
Rank-Wise Mixture-of-ExpertsOrganizations
None yet
submitted a
paper to Daily Papers 30 days ago
authored 10
papers 30 days ago
Fly-CL: A Fly-Inspired Framework for Enhancing Efficient Decorrelation and Reduced Training Time in Pre-trained Model-based Continual Representation Learning
Paper • 2510.16877 • Published • 4
FlyLoRA: Boosting Task Decoupling and Parameter Efficiency via Implicit Rank-Wise Mixture-of-Experts
Paper • 2510.08396 • Published • 5
Utility-Diversity Aware Online Batch Selection for LLM Supervised Fine-tuning
Paper • 2510.16882 • Published • 5
Enhancing Pretrained Model-based Continual Representation Learning via Guided Random Projection
Paper • 2603.19145 • Published
Small Generalizable Prompt Predictive Models Can Steer Efficient RL Post-Training of Large Reasoning Models
Paper • 2602.01970 • Published • 5
Energy-based Automated Model Evaluation
Paper • 2401.12689 • Published • 1
Listwise Policy Optimization: Group-based RLVR as Target-Projection on the LLM Response Simplex
Paper • 2605.06139 • Published • 69
Dynamics-Predictive Sampling for Active RL Finetuning of Large Reasoning Models
Paper • 2603.10887 • Published
Structural features of the fly olfactory circuit mitigate the stability-plasticity dilemma in continual learning
Paper • 2502.01427 • Published • 1
TRACE: A Unified Rollout Budget Allocation Framework for Efficient Agentic Reinforcement Learning
Paper • 2606.11119 • Published • 18
upvoted 2
papers about 1 month ago
upvoted a paper 2 months ago
upvoted 3
papers 9 months ago
Utility-Diversity Aware Online Batch Selection for LLM Supervised Fine-tuning
Paper • 2510.16882 • Published • 5
Fly-CL: A Fly-Inspired Framework for Enhancing Efficient Decorrelation and Reduced Training Time in Pre-trained Model-based Continual Representation Learning
Paper • 2510.16877 • Published • 4
FlyLoRA: Boosting Task Decoupling and Parameter Efficiency via Implicit Rank-Wise Mixture-of-Experts
Paper • 2510.08396 • Published • 5