# JarvisBench: Always-on Intelligence Between Humans and Agents

arXiv: https://arxiv.org/abs/2608.14870 · HTML: https://mlsanity.com/paper/2608.14870/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 9 of its 14 references that are in the map.
- First posted on arXiv 14 Aug 2026.

## Builds on (most cited first, 9 of 9)

- [AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversation](https://mlsanity.com/paper/2308.08155/) (2023)
- [MetaGPT: Meta Programming for A Multi-Agent Collaborative Framework](https://mlsanity.com/paper/2308.00352/) (2023)
- [τ-bench: A Benchmark for Tool-Agent-User Interaction in Real-World Domains](https://mlsanity.com/paper/2406.12045/) (2024)
- [MultiAgentBench: Evaluating the Collaboration and Competition of LLM agents](https://mlsanity.com/paper/2503.01935/) (2025)
- [Qwen3-ASR Technical Report](https://mlsanity.com/paper/2601.21337/) (2026)
- [UserBench: An Interactive Gym Environment for User-Centric Agents](https://mlsanity.com/paper/2507.22034/) (2025)
- [VitaBench: Benchmarking LLM Agents with Versatile Interactive Tasks in Real-world Applications](https://mlsanity.com/paper/2509.26490/) (2025)
- [SoulX-Duplug: Plug-and-Play Streaming State Prediction Module for Realtime Full-Duplex Speech Conversation](https://mlsanity.com/paper/2603.14877/) (2026)
- [π-Bench: Evaluating Proactive Personal Assistant Agents in Long-Horizon Workflows](https://mlsanity.com/paper/2605.14678/) (2026)

## Abstract (arXiv)

Long-horizon agents can execute continuously, but human attention remains intermittent and scarce. This creates a bidirectional coordination problem: users may need immediate access to an agent while work continues in the background, whereas agents may encounter consequential decisions that require user judgment after the user has stopped monitoring execution. We posit an always-on attention-coordination layer---textitJarvisfootnoteNamed after the fictional AI assistant in textitIron Man.---that mediates this interface and allocates human attention across one or more working agents. We introduce textitJarvisBench to evaluate both directions of this coordination: whether an intermediary can accurately and promptly answer user-initiated questions about ongoing work, and whether it can recognize when an agent requires user judgment, solicit that judgment at the right moment, and route it back to improve task outcomes. JarvisBench contains 45 agentic task instances: 20 single-agent tasks and 25 workstreams organized into 10 multi-agent projects. The tasks span 19 domains and were selected and adapted from more than 2,000 public candidates. Crucially, the need for user attention arises naturally during execution rather than from an obvious omission in the initial prompt. JarvisBench is designed to integrate with arbitrary agent runtimes without modifying their underlying execution loops. Our reference implementation further provides a full-duplex speech interface, allowing users to reach Jarvis naturally while timely attention coordination supports agents working in the background. By separating agent execution from attention coordination, JarvisBench provides a stable evaluation target as agent capabilities continue to improve.