PIRA-Bench: A Transition from Reactive GUI Agents to GUI-based Proactive Intent Recommendation Agents

TL;DR

This paper introduces PIRA-Bench, a new benchmark for evaluating multimodal large language models in proactive GUI-based assistant tasks, addressing challenges of real-world, noisy, and complex visual input trajectories.

Contribution

The paper presents PIRA-Bench, a novel benchmark for proactive GUI agents, and proposes PIRF, a baseline framework for managing multiple tasks and noisy inputs in visual assistant scenarios.

Findings

PIRA-Bench effectively challenges models with complex, noisy visual trajectories.

PIRF baseline demonstrates improved task management and robustness in proactive GUI tasks.

Benchmark paves the way for developing more intelligent, anticipatory GUI assistants.

Abstract

Current Graphical User Interface (GUI) agents operate primarily under a reactive paradigm: a user must provide an explicit instruction for the agent to execute a task. However, an intelligent AI assistant should be proactive, which is capable of anticipating user intentions directly from continuous visual inputs, such as mobile or desktop screenshots, and offering timely recommendations without explicit user prompting. Transitioning to this proactive paradigm presents significant challenges. Real-world screen activity is rarely linear; it consists of long-horizon trajectories fraught with noisy browsing, meaningless actions, and multithreaded task-switching. To address this gap, we introduce PIRA-Bench (Proactive Intent Recommendation Agent Benchmark), a novel benchmark for evaluating multimodal large language models (MLLMs) on continuous, weakly-supervised visual inputs. Unlike reactive datasets, PIRA-Bench features complex trajectories with multiple interleaved intents and noisy segments with various user profile contexts, challenging agents to detect actionable events while fitting to user preferences. Furthermore, we propose the PIRF baseline, a memory-aware, state-tracking framework that empowers general MLLMs to manage multiple task threads and handle misleading visual inputs. PIRA-Bench serves as an initial step toward robust and proactive GUI-based personal assistants.