VibeLifeBench: Can Your Life Agent Be Proactive and Persistent in a Living World?

Source

• Source name: arXiv
• Source kind: paper
• Authors / organization: Xiaohongshu Inc
• Venue: arXiv: cs.CL
• Published: 2026-08-11
• Collected: 2026-08-12
• URL: http://arxiv.org/abs/2608.10875v1
• PDF: https://arxiv.org/pdf/2608.10875v1
• DOI: UNKNOWN
• Registry group: academic_apis_and_preprints
• Evidence role: canonical_academic_index
• Language: en
• Country / region: UNKNOWN

Plain-Language Brief

The paper introduces VibeLifeBench, a new benchmark for evaluating large language model (LLM) agents' ability to handle long-horizon tasks in everyday life scenarios. The benchmark consists of 200 multi-week timeline tasks across ten everyday-life domains simulating 22 mock services. The world progresses independently and silently, requiring an agent that re-inspects the world to discover changes. The evaluation grades agents based on end state, timeliness of actions, and adherence to implicit constraints.

What the Document Says

• The text discusses the need for a new benchmark (VibeLifeBench) to evaluate LLM agents' performance in long-horizon tasks of everyday life assistance.
• The benchmark consists of 200 multi-week tasks across ten everyday-life domains simulating 22 mock services.
• The world progresses on its own clock, and many changes are silent, requiring an agent that re-inspects the world to discover them.
• Evaluation is based on fine-grained, weighted checks reading only what the agent left behind.
• Seven frontier models were evaluated with all scoring low, showing current agents' limitations in assisting with real life.

Key Concepts Mentioned

• Large language model (LLM) agents
• Everyday life assistance
• Long-horizon tasks
• VibeLifeBench
• Evaluation framework

Claims Made by the Document

• Current LLM agents fail at long-term everyday life assistance.
• Existing evaluations are insufficient for measuring proactive and persistent agents.

Evidence Used

• Low scores of seven frontier models in the new benchmark.

Methods / Approach

UNKNOWN

Limitations Stated by the Authors

UNKNOWN

Limitations Noted by the Agent

Agent interpretation — not from the source.

UNKNOWN

Open Questions

UNKNOWN

Notes for Future Q&A

May include interpretation. Labeled where it does.

UNKNOWN

Raw Abstract / Excerpt

Large language model (LLM) agents are increasingly deployed as personal assistants. Existing evaluations, however, mostly use short, self-contained requests in static environments. Everyday life assistance is different. A task runs for weeks rather than minutes. The world keeps changing while the agent is not being prompted. Many constraints are never stated outright. An agent that merely answers the request in front of it will fail at such a task. What is needed instead is an agent that stays proactive and consistent. It decides on its own when to act, when to ask, and when to stay silent. It notices changes that nobody announced. It keeps one plan coherent from the first day to the last. No current benchmark measures this. We introduce VibeLifeBench, a benchmark of 200 long-horizon tasks across ten everyday-life domains. Each task is a scripted multi-week timeline in a simulated world of 22 mock services. The world advances on its own clock, and many of its changes are silent, so only an agent that re-inspects the world discovers them. Every task is graded by fine-grained, weighted checks that read only what the agent actually left behind, covering the end state, the timeliness of its actions, and whether it upheld the implicit constraints. We evaluate seven frontier models. All of them score low, which shows how far current agents are from assisting with real life. We will open-source all tasks, environments, and the evaluation framework.