Rethinking Training-Inference Mismatch in LLM Reinforcement Learning: Where It Arises and How to Correct It
Paper • 2609.32444 • Published • 35
Kaixiang Zhao
kzhao5
AI & ML interests
None yet
Recent Activity
authored a paper 13 days ago
Rethinking Training-Inference Mismatch in LLM Reinforcement Learning: Where It Arises and How to Correct It upvoted a paper 13 days ago
An RL View of OPD: Least Square Policy Distillation for Sample-Efficient LLM ReasoningOrganizations
None yet
authored a
paper 13 days ago
upvoted 2
papers 13 days ago
submitted a
paper to Daily Papers 13 days ago
updated a
model 17 days ago
updated 3
models 21 days ago
published a
model 21 days ago
upvoted a paper 23 days ago
authored a
paper 24 days ago
upvoted a paper 24 days ago
submitted a
paper to Daily Papers 24 days ago
authored 7
papers 24 days ago
A Survey of Model Extraction Attacks and Defenses in Distributed Computing Environments
Paper • 2502.16065 • Published
DESIGN: Encrypted GNN Inference via Server-Side Input Graph Pruning
Paper • 2507.05649 • Published
A Systematic Survey of Model Extraction Attacks and Defenses: State-of-the-Art and Perspectives
Paper • 2508.15031 • Published
Intellectual Property in Graph-Based Machine Learning as a Service: Attacks and Defenses
Paper • 2508.19641 • Published
Provable and Practical In-Context Policy Optimization for Self-Improvement
Paper • 2603.01335 • Published
GraphIP-Bench: How Hard Is It to Steal a Graph Neural Network, and Can We Stop It?
Paper • 2605.12827 • Published
FRAME: Forensic Routing and Adaptive Multi-path Evidence Fusion for Image Manipulation Detection
Paper • 2605.12826 • Published