# Ask Now, Use Later: Benchmarking the Proactivity Gap in Long-Lived LLM Agents

arXiv: https://arxiv.org/abs/2605.28108 · HTML: https://mlsanity.com/paper/2605.28108/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Placed in General LLM agent frameworks and benchmarks, one of 16 directions in LLM agents and tool use.
- Builds on 20 of its 40 references that are in the map.
- First posted on arXiv 27 May 2026.

## Placed in

- LLM agents and tool use / [General LLM agent frameworks and benchmarks](https://mlsanity.com/agents/general-llm-agent-frameworks-and-benchmarks/)

## Builds on (most cited first, 20 of 20)

- [Generative Agents: Interactive Simulacra of Human Behavior](https://mlsanity.com/paper/2304.03442/) (2023)
- [τ-bench: A Benchmark for Tool-Agent-User Interaction in Real-World Domains](https://mlsanity.com/paper/2406.12045/) (2024)
- [MemGPT: Towards LLMs as Operating Systems](https://mlsanity.com/paper/2310.08560/) (2023)
- [DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence](https://mlsanity.com/paper/2606.19348/) (2026)
- [Evaluating Very Long-Term Conversational Memory of LLM Agents](https://mlsanity.com/paper/2402.17753/) (2024)
- [LaMP: When Large Language Models Meet Personalization](https://mlsanity.com/paper/2304.11406/) (2023)
- [Do LLMs Recognize Your Preferences? Evaluating Personalized Preference Following in LLMs](https://mlsanity.com/paper/2502.09597/) (2025)
- [Eliciting Human Preferences with Language Models](https://mlsanity.com/paper/2310.11589/) (2023)
- [The MiniMax-M2 Series: Mini Activations Unleashing Max Real-World Intelligence](https://mlsanity.com/paper/2605.26494/) (2026)
- [Ask-before-Plan: Proactive Language Agents for Real-World Planning](https://mlsanity.com/paper/2406.12639/) (2024)
- [Inferring Rewards from Language in Context](https://mlsanity.com/paper/2204.02515/) (2022)
- [BED-LLM: Intelligent Information Gathering with LLMs and Bayesian Experimental Design](https://mlsanity.com/paper/2508.21184/) (2025)
- [KnowU-Bench: Towards Interactive, Proactive, and Personalized Mobile Agent Evaluation](https://mlsanity.com/paper/2604.08455/) (2026)
- [PersonalAlign: Hierarchical Implicit Intent Alignment for Personalized GUI Agent with Long-Term User-Centric Records](https://mlsanity.com/paper/2601.09636/) (2026)
- [Proactive Agent Research Environment: Simulating Active Users to Evaluate Proactive Assistants](https://mlsanity.com/paper/2604.00842/) (2026)
- [VitaBench 2.0: Evaluating Personalized and Proactive Agents in Long-Term User Interactions](https://mlsanity.com/paper/2605.27141/) (2026)
- [LifeSim: Long-Horizon User Life Simulator for Personalized Assistant Evaluation](https://mlsanity.com/paper/2603.12152/) (2026)
- [PIRA-Bench: A Transition from Reactive GUI Agents to GUI-based Proactive Intent Recommendation Agents](https://mlsanity.com/paper/2603.08013/) (2026)
- [π-Bench: Evaluating Proactive Personal Assistant Agents in Long-Horizon Workflows](https://mlsanity.com/paper/2605.14678/) (2026)
- [Springdrift: An Auditable Persistent Runtime for LLM Agents with Case-Based Memory, Normative Safety, and Ambient Self-Perception](https://mlsanity.com/paper/2604.04660/) (2026)

## Built on by (most cited first, 2 of 2)

- [Agents in the Large: Perception-Centered Architecture for Persistent Agents](https://mlsanity.com/paper/2608.30478/) (2026)
- [Asking for What Was Never Requested: Horizontal and Vertical Proactivity in Agents](https://mlsanity.com/paper/2609.37236/) (2026)

## Abstract (arXiv)

A long-lived LLM agent, such as OpenClaw, earns its value by acting on a user's preferences and constraints across sessions, not just the current request. Yet today's agents keep what a user volunteers but rarely ask for what stays unspoken, leaving a proactivity gap in long-lived LLM agents: an agent cannot act on a preference it never obtained. As users delegate more of their affairs to agents, the impact of this gap grows. We isolate one concrete, controllable slice of this gap as Ask-to-Remember (ATR): the agent decides whether to ask now for a reusable user preference that the current task does not need but a later session with the same user will. ATR is hard even to evaluate: the right question is underdetermined and its payoff deferred to tasks that may never arise. ATRBench, to the best of our knowledge the first ATR benchmark, makes it measurable by fixing each user's preferences as hidden ground truth, so success demands asking, not recall. Across eight frontier LLM agents, defaults fall at least 62 points below an oracle handed the relevant preference, and prompting closes little of it. Diagnostics identify acquisition as the bottleneck. ATRBench surfaces this proactivity gap in current agents and offers a diagnostic testbed for closing it.