Act Wisely: Cultivating Meta-Cognitive Tool Use in Agentic Multimodal Models Paper • 2604.08545 • Published 4 days ago • 36
MM-CondChain: A Programmatically Verified Benchmark for Visually Grounded Deep Compositional Reasoning Paper • 2603.12266 • Published Mar 12 • 19
SwimBird: Eliciting Switchable Reasoning Mode in Hybrid Autoregressive MLLMs Paper • 2602.06040 • Published Feb 5 • 10