Beyond Correctness: Benchmarking and Aligning Response Behaviors in Hybrid-Thinking MLLMs
Paper • 2608.12781 • Published • 35
周文睿
William422
AI & ML interests
None yet
Recent Activity
upvoted a paper about 2 months ago
Beyond Correctness: Benchmarking and Aligning Response Behaviors in Hybrid-Thinking MLLMs upvoted a paper 2 months ago
PCSD: Persistent Consistency for Self-Distillation in Agentic Reinforcement Learning upvoted a paper 12 months ago
AdaSPEC: Selective Knowledge Distillation for Efficient Speculative
DecodersOrganizations
None yet