CORPGEN: Simulating Corporate Environments with Autonomous Digital Employees in Multi-Horizon Task Environments

TL;DR

This paper introduces Multi-Horizon Task Environments (MHTEs) to evaluate autonomous agents on managing multiple complex, interleaved long-term tasks, and presents CorpGen, a framework that improves agent performance and stability in these settings.

Contribution

We propose MHTEs as a new benchmark for long-horizon, multi-task reasoning and develop CorpGen, a novel architecture addressing key failure modes in complex task management.

Findings

CorpGen achieves up to 3.5x performance improvement over baselines.

Performance remains stable under increasing task load.

Ablation shows experiential learning yields significant gains.

Abstract

Long-horizon reasoning is a key challenge for autonomous agents, yet existing benchmarks evaluate agents on single tasks in isolation. Real organizational work requires managing many concurrent long-horizon tasks with interleaving, dependencies, and reprioritization. We introduce Multi-Horizon Task Environments (MHTEs): a distinct problem class requiring coherent execution across dozens of interleaved tasks (45+, 500-1500+ steps) within persistent execution contexts spanning hours. We identify four failure modes that cause baseline CUAs to degrade from 16.7% to 8.7% completion as load scales 25% to 100%, a pattern consistent across three independent implementations. These failure modes are context saturation (O(N) vs O(1) growth), memory interference, dependency complexity (DAGs vs. chains), and reprioritization overhead. We present CorpGen, an architecture-agnostic framework addressing these failures via hierarchical planning for multi-horizon goal alignment, sub-agent isolation preventing cross-task contamination, tiered memory (working, structured, semantic), and adaptive summarization. CorpGen simulates corporate environments through digital employees with persistent identities and realistic schedules. Across three CUA backends (UFO2, OpenAI CUA, hierarchical) on OSWorld Office, CorpGen achieves up to 3.5x improvement over baselines (15.2% vs 4.3%) with stable performance under increasing load, confirming that gains stem from architectural mechanisms rather than specific CUA implementations. Ablation studies show experiential learning provides the largest gains.