5 papers across 3 sessions
MM-OPERA is a benchmark of 11,497 open-ended association tasks (Remote-Item and In-Context) with explicit multi-hop reasoning and LLM-as-a-Judge process-reward evaluation to assess LVLMs' convergent and divergent associative thinking.
To mitigate temporal redundancy for SNNs, we propose a module named Mutual Information-based Temporal Redundancy Quantification and Reduction (MI-TRQR), constructing energy-efficient SNNs.
