Abstract: Recent robot task planners utilize large language models (LLMs) or vision-language models (VLMs) as a failure detector. These methods perform well by leveraging their semantic reasoning ...
GDF-MIL (Graph-Driven Fusion Multiple Instance Learning) is a novel graph-driven multi-instance learning framework that adaptively balances topology modeling and semantic feature preservation through ...
Abstract: Previous multi-agent deep reinforcement learning (MADRL) algorithms have shown strong performance in symmetric scenarios. However, research on numerically disadvantaged scenarios is limited ...