JARVIS-1
First page
Paper summary
An open-world multimodal agent for Minecraft that combines perception, planning, and memory into a self-improving system.
Ask this paper
01
Multimodal perception: Processes visual Minecraft observations and natural-language instructions through a unified multimodal input pipeline.
02
Memory-augmented planning: Maintains a multimodal memory store of past observations and plans, enabling lifelong self-improvement across episodes.
03
Strong task coverage: Completes 200+ diverse Minecraft tasks with competitive success rates, including long-horizon tasks like diamond collection.
04
Open-world blueprint: An influential example of combining foundation models, memory, and explicit planning into an agent, foreshadowing many 2024 agent architectures.