VIKI‑R: Coordinating Embodied Multi-Agent Cooperation via Reinforcement Learning

Li Kang (Shanghai Jiaotong University) · Xiufeng Song (Shanghai Jiao Tong University) · Heng Zhou (University of Science and Technology of China) · Yiran Qin (The Chinese University of Hong Kong(Shenzhen)) · Jie Yang (University of Illinois at Chicago) · Xiaohong Liu (Shanghai Jiao Tong University) · Philip Torr (University of Oxford) · LEI BAI (UNSW, Sydney) · Zhenfei Yin (University of Oxford)

agent activationchain-of-thoughtcompositional cooperation patternsdynamic environments embodied agentshierarchical benchmarkmulti-view visual observationsperception-driven reasoningreinforcement learningscalable cooperation strategiesstructured supervision signalstask planningtrajectory perceptionvision-language models visual reasoning

Coordinating multiple embodied agents in dynamic environments remains a core challenge in artificial intelligence, requiring both perception-driven reasoning and scalable cooperation strategies. While recent works have leveraged large language models (LLMs) for multi-agent planning, a few have begun to explore vision-language models (VLMs) for visual reasoning. However, these VLM-based approaches remain limited in their support for diverse embodiment types. In this work, we introduce VIKI-Bench, the first hierarchical benchmark tailored for embodied multi-agent cooperation, featuring three structured levels: agent activation, task planning, and trajectory perception. VIKI-Bench includes diverse robot embodiments, multi-view visual observations, and structured supervision signals to evaluate reasoning grounded in visual inputs. To demonstrate the utility of VIKI-Bench, we propose VIKI-R, a two-stage framework that fine-tunes a pretrained vision-language model (VLM) using Chain-of-Thought annotated demonstrations, followed by reinforcement learning under multi-level reward signals. Our extensive experiments show that VIKI-R significantly outperforms baselines method across all task levels. Furthermore, we show that reinforcement learning enables the emergence of compositional cooperation patterns among heterogeneous agents. Together, VIKI-Bench and VIKI-R offer a unified testbed and method for advancing multi-agent, visual-driven cooperation in embodied AI systems.