Undergrad student, SUN YAT-SEN UNIVERSITY
2 papers at NeurIPS 2025
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.
We propose GAM-Agent, a game-theoretic multi-agent framework where visual and logic agents debate via structured communication and uncertainty control, boosting VLM performance, robustness, and interpretability. It is modular, scalable, and general.
