OSExpert: Computer-Use Agents Learning Professional Skills via Exploration

TL;DR

This paper introduces OSExpert, a learning agent for computer use tasks that employs a novel exploration algorithm and skill composition to improve performance and efficiency, approaching human expert levels.

Contribution

The paper presents a GUI-based exploration algorithm and a skill self-construction method that significantly enhance agent performance in complex digital tasks.

Findings

Achieves 20% performance improvement on OSExpert-Eval

Reduces inference-time scaling inefficiency by 80%

Demonstrates effective skill transfer and compositionality

Abstract

General-purpose computer-use agents have shown impressive performance across diverse digital environments. However, our new benchmark, OSExpert-Eval, indicates they remain far less helpful than human experts. Although inference-time scaling enables adaptation, these agents complete complex tasks inefficiently with degraded performance, transfer poorly to unseen UIs, and struggle with fine-grained action sequences. To solve the problem, we introduce a GUI-based depth-first search (GUI-DFS) exploration algorithm to comprehensively explore and verify an environment's unit functions. The agent then exploits compositionality between unit skills to self-construct a curriculum for composite tasks. To support fine-grained actions, we curate a database of action primitives for agents to discover during exploration; these are saved as a skill set once the exploration is complete. We use the learned skills to improve the agent's performance and efficiency by (1) enriching agents with ready-to-use procedural knowledge, allowing them to plan only once for long trajectories and generate accurate actions, and (2) enabling them to end inference-time scaling earlier by realizing their boundary of capabilities. Extensive experiments show that our environment-learned agent takes a meaningful step toward expert-level computer use, achieving a around 20 percent performance gain on OSExpert-Eval and closing the efficiency gap to humans by around 80 percent