Pith. sign in

REVIEW 13 cited by

JARVIS-1: Open-World Multi-task Agents with Memory-Augmented Multimodal Language Models

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2311.05997 v3 pith:NUSO4L65 submitted 2023-11-10 cs.AI

classification cs.AI
keywords jarvis-1tasksmultimodalopen-worldagentscontrolobservationsplans
verification ladder T0 review T1 audit T2 compute T3 formal

Signed reviews

No signed human review yet.

0 comments
abstract

Achieving human-like planning and control with multimodal observations in an open world is a key milestone for more functional generalist agents. Existing approaches can handle certain long-horizon tasks in an open world. However, they still struggle when the number of open-world tasks could potentially be infinite and lack the capability to progressively enhance task completion as game time progresses. We introduce JARVIS-1, an open-world agent that can perceive multimodal input (visual observations and human instructions), generate sophisticated plans, and perform embodied control, all within the popular yet challenging open-world Minecraft universe. Specifically, we develop JARVIS-1 on top of pre-trained multimodal language models, which map visual observations and textual instructions to plans. The plans will be ultimately dispatched to the goal-conditioned controllers. We outfit JARVIS-1 with a multimodal memory, which facilitates planning using both pre-trained knowledge and its actual game survival experiences. JARVIS-1 is the existing most general agent in Minecraft, capable of completing over 200 different tasks using control and observation space similar to humans. These tasks range from short-horizon tasks, e.g., "chopping trees" to long-horizon tasks, e.g., "obtaining a diamond pickaxe". JARVIS-1 performs exceptionally well in short-horizon tasks, achieving nearly perfect performance. In the classic long-term task of $\texttt{ObtainDiamondPickaxe}$, JARVIS-1 surpasses the reliability of current state-of-the-art agents by 5 times and can successfully complete longer-horizon and more challenging tasks. The project page is available at https://craftjarvis.org/JARVIS-1

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 13 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. TeamCraft: A Benchmark for Multi-Modal Multi-Agent Systems in Minecraft

    cs.AI 2024-12 conditional novelty 7.0 of 10

    TeamCraft presents a large multi-modal, multi-agent Minecraft benchmark and shows that current models generalize poorly to novel goals, scenes, and team sizes.

  2. Weights or Skills? A Survey of Robot-Learning Techniques: from Action-Predicting Weights to Robots that Write their Own Skills

    cs.RO 2026-08 conditional novelty 6.0 of 10

    A taxonomy of robot learning on a weights-versus-skills axis, with a five-rung self-improvement ladder whose top cell (feedback plus memory plus search) holds only a few recent systems.

  3. HiMe: Hierarchical Embodied Memory for Long-Horizon Vision-Language-Action Control

    cs.RO 2026-07 conditional novelty 6.0 of 10

    A three-layer embodied memory (Executor, Sentry, Planner) with cross-modal Add/Update/Delete operations raises long-horizon VLA task progress to ~90% and reduces planner API calls versus flat memory.

  4. CrafterDojo: A Suite of Foundation Models for Building Open-Ended Embodied Agents in Crafter

    cs.AI 2025-08 unverdicted novelty 6.0 of 10

    CrafterDojo provides CrafterVPT, CrafterCLIP, and CrafterSteve-1 models plus datasets and benchmarks, making Crafter a usable testbed for general-purpose embodied agents.

  5. Conditional Multi-Stage Failure Recovery for Embodied Agents

    cs.CL 2025-07 conditional novelty 6.0 of 10

    A conditional four-stage chain-prompting method for failure recovery improves success on the TEACH embodied-agent benchmark from 24.9% to 36.5% with the same plan and executor.

  6. Embodied VideoAgent: Persistent Memory from Egocentric Videos and Embodied Sensors Enables Dynamic Scene Understanding

    cs.CV 2024-12 conditional novelty 6.0 of 10

    Embodied VideoAgent augments an LLM-based video agent with persistent object memory built from egocentric video, depth, and pose, plus VLM-based memory updates, reporting gains on Ego4D-VQ3D, OpenEQA, and EnvQA.

  7. Plancraft: an evaluation dataset for planning with LLM agents

    cs.CL 2024-12 conditional novelty 6.0 of 10

    A new Minecraft crafting benchmark shows current LLM agents solve only a minority of planning tasks, and vision-language models almost never succeed on image observations.

  8. MineStudio: A Streamlined Package for Minecraft AI Agent Development

    cs.AI 2024-12 conditional novelty 5.0 of 10

    A software framework that integrates seven engineering components to streamline Minecraft AI agent development.

  9. GROOT-2: Weakly Supervised Multi-Modal Instruction Following Agents

    cs.AI 2024-12 conditional novelty 5.0 of 10

    A weakly supervised latent-variable agent improves multimodal instruction following by combining VAE self-imitating on unlabeled data with a likelihood-based alignment of labeled and video latents.

  10. Visual Language Models as Operator Agents in the Space Domain

    cs.AI 2025-01 conditional novelty 4.0 of 10

    Vision-language models can act as spacecraft operators in the KSPDG simulator from screenshots, and fine-tuning OpenVLA on ten episodes shows preliminary promise for robotic satellite inspection.

  11. FaGeL: Fabric LLMs Agent empowered Embodied Intelligence Evolution with Autonomous Human-Machine Collaboration

    cs.HC 2024-12 reject novelty 4.0 of 10

    A smart-fabric LLM agent that learns from textual feedback via a token-level DualCUT alignment method, with an 11.3% Overcooked-AI score gain claimed.

  12. WiS Platform: Enhancing Evaluation of LLM-Based Multi-Agent Systems Through Game-Based Analysis

    cs.AI 2024-12 conditional novelty 4.0 of 10

    An open platform and benchmark for evaluating LLM-based multi-agent systems through the 'Who is Spy?' game, including a leaderboard and behavioral analysis of ten models.

  13. STEVE-Audio: Expanding the Goal Conditioning Modalities of Embodied Agents in Minecraft

    cs.LG 2024-12 conditional novelty 4.0 of 10

    An audio-conditioned STEVE-1 agent, built with a new Minecraft audio-video CLIP model and a learned prior, matches or beats text- and video-conditioned versions on most short-horizon collection tasks.

Pith tools