Anticipate and Learn: Unleashing Idle-Time Compute in Proactive Agents

• URL: http://arxiv.org/abs/2605.25971v2
• Date: Tue, 26 May 2026 04:49:24 GMT
• Title: Anticipate and Learn: Unleashing Idle-Time Compute in Proactive Agents
• Abstract summary: ProAct is a proactive agent architecture that leverages idle-time compute to anticipate and fulfill likely upcoming user needs.<n>By analyzing evolving dialogue history together with persistent memory, ProAct predicts upcoming needs and iteratively acquires information.<n>MemBench evaluations confirm that ProAct achieves state-of-the-art reflective accuracy, underscoring its sustained and robust performance.
• Score: 51.46570338144688
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: While AI agents demonstrate remarkable capabilities in reasoning and tool use, they remain fundamentally reactive: they compute responses only after explicit user prompts. This paradigm ignores a critical opportunity: the idle time between interactions is largely wasted, leaving agents unable to prepare for future user needs. To bridge this gap, we introduce ProAct, a proactive agent architecture that leverages idle-time compute to anticipate and fulfill likely upcoming user needs. By analyzing evolving dialogue history together with persistent memory, ProAct predicts upcoming needs and iteratively acquires information, allowing the agent to resolve knowledge gaps and prepare evidence before the user initiates a query. To rigorously evaluate proactive capabilities, we also introduce ProActEval, a comprehensive benchmark comprising 200 scenarios across 40 domains, featuring predictable need chains and diverse user cognitive profiles. Empirical results demonstrate significant advantages over reactive baselines. ProAct accelerates task completion by reducing required turns by 14.8%, decreases user effort by 11.7%, and cuts hallucination rates by 28.1% on ProActEval. Furthermore, MemBench evaluations confirm that ProAct achieves state-of-the-art reflective accuracy, underscoring its sustained and robust performance.

Related papers

• Act2Intention: A Benchmark For Developing Active Mobile Agents Through Inferring User Intention from GUI Actions [18.497151729519267]
  We propose the Act2Intention framework that builds an active mobile agent by integrating understanding, predicting user intentions, and executing decisions.<n>We construct the Act2Intention Bench through data collection and validated generation, comprising 72,511 intentions and over 700,000 actions across 52 apps.<n> Experimental results show that supervised fine-tuning on Act2Intention Bench yields absolute improvements of +32.0 Acc-S, +10.25 Acc-S, and +6.9 SSR points over non-fine-tuned counterparts.
  arXiv  Detail & Related papers  (2026-08-14T09:36:59Z)
• ProEvent: An Event-centric Benchmark for Proactive Agents [41.03981463319731]
  ProEvent is the first event-centric benchmark designed to assess an agent's ability to proactively maintain a user's timetable.<n>ProEvent provides synthesized yet realistic chats that consider the dynamic interaction among users, concurrent chat threads, and noise in the real world.
  arXiv  Detail & Related papers  (2026-07-20T08:48:36Z)
• ProactiveBench: Benchmarking Proactiveness in Multimodal Large Language Models [51.06179253173491]
  We introduce ProactiveBench, a benchmark that tests proactiveness across different tasks.<n>We evaluate 22 MLLMs on ProactiveBench, showing that (i) they generally lack proactiveness; (ii) proactiveness does not correlate with model capacity; (iii) "hinting" at proactiveness yields only marginal gains.
  arXiv  Detail & Related papers  (2026-03-19T20:58:03Z)
• ProactiveMobile: A Comprehensive Benchmark for Boosting Proactive Intelligence on Mobile Devices [17.39388308538324]
  This paper introduces ProactiveMobile, a benchmark for proactive mobile agent development.<n>It formalizes the proactive task as inferring latent user intent across four dimensions of on-device contextual signals.<n>The benchmark achieves a success rate of 19.15%, outperforming o1 (15.71%) and GPT-5 (7.39%) in experiments.
  arXiv  Detail & Related papers  (2026-02-25T12:32:37Z)
• Pushing Forward Pareto Frontiers of Proactive Agents with Behavioral Agentic Optimization [61.641777037967366]
  Proactive large language model (LLM) agents aim to actively plan, query, and interact over multiple turns.<n>Agentic reinforcement learning (RL) has emerged as a promising solution for training such agents in multi-turn settings.<n>We propose BAO, an agentic RL framework that combines behavior enhancement to enrich proactive reasoning and information-gathering capabilities.
  arXiv  Detail & Related papers  (2026-02-11T20:40:43Z)
• ProAgent: Harnessing On-Demand Sensory Contexts for Proactive LLM Agent Systems [7.591337469415894]
  ProAgent is an end-to-end proactive agent system that harnesses massive sensory contexts and LLM reasoning to deliver proactive assistance.<n>We implement ProAgent on Augmented Reality (AR) glasses with an edge server and extensively evaluate it on a real-world testbed, a public dataset, and through a user study.<n>Results show that ProAgent achieves up to 33.4% higher proactive prediction accuracy, 16.8% higher tool-calling F1 score, and notable improvements in user satisfaction.
  arXiv  Detail & Related papers  (2025-12-07T08:21:07Z)
• Active Test-time Vision-Language Navigation [60.69722522420299]
  ATENA is a test-time active learning framework that enables a practical human-robot interaction via episodic feedback on uncertain navigation outcomes.<n>In particular, ATENA learns to increase certainty in successful episodes and decrease it in failed ones, improving uncertainty calibration.<n>In addition, we propose a self-active learning strategy that enables an agent to evaluate its navigation outcomes based on confident predictions.
  arXiv  Detail & Related papers  (2025-06-07T02:24:44Z)
• Proactive Agent: Shifting LLM Agents from Reactive Responses to Active Assistance [95.03771007780976]
  We tackle the challenge of developing proactive agents capable of anticipating and initiating tasks without explicit human instructions.<n>First, we collect real-world human activities to generate proactive task predictions.<n>These predictions are labeled by human annotators as either accepted or rejected.<n>The labeled data is used to train a reward model that simulates human judgment.
  arXiv  Detail & Related papers  (2024-10-16T08:24:09Z)
• Beyond Trial-and-Error: Predicting User Abandonment After a Moderation Intervention [0.6918368994425961]
  Current content moderation follows a reactive, trial-and-error approach.<n>We introduce a proactive, predictive approach that enables moderators to anticipate the impact of their actions before implementation.<n>We study the reactions of 16,540 users to a massive ban of online communities on Reddit.
  arXiv  Detail & Related papers  (2024-04-23T08:52:41Z)
• Augmenting Unsupervised Reinforcement Learning with Self-Reference [63.68018737038331]
  Humans possess the ability to draw on past experiences explicitly when learning new tasks. We propose the Self-Reference (SR) approach, an add-on module explicitly designed to leverage historical information. Our approach achieves state-of-the-art results in terms of Interquartile Mean (IQM) performance and Optimality Gap reduction on the Unsupervised Reinforcement Learning Benchmark.
  arXiv  Detail & Related papers  (2023-11-16T09:07:34Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.