COBRA-Skills: Contextual Bandit-Guided Evolution for Agent Skill Optimization
Paper • 2609.11682 • Published • 46
Zhongxiang Dai
dzxagent
AI & ML interests
Agents, LLMs
Recent Activity
upvoted a paper 11 days ago
COBRA-Skills: Contextual Bandit-Guided Evolution for Agent Skill Optimization submitted a paper about 1 month ago
SPOT: Sparse Probing and Outcome Calibration for On-Policy Distillation authored a paper about 2 months ago
SPOT: Sparse Probing and Outcome Calibration for On-Policy DistillationOrganizations
None yet