Beyond Correctness: Benchmarking and Aligning Response Behaviors in Hybrid-Thinking MLLMs
Paper • 2608.12781 • Published • 25 •
4
Matan Avitan
MatanAvitan
AI & ML interests
None yet
Recent Activity
commentedon a paper about 11 hours ago
Beyond Correctness: Benchmarking and Aligning Response Behaviors in Hybrid-Thinking MLLMs authored a paper about 2 months ago
Power-Softmax: Towards Secure LLM Inference over Encrypted Data authored a paper about 2 months ago
MANCE: Manifold Aware Concept ErasureOrganizations
None yet
commented a
paper about 11 hours ago