CORPGEN: Simulating Corporate Environments with Autonomous Digital Employees in Multi-Horizon Task Environments
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866908835437871104 |
|---|---|
| author | Jaye, Abubakarr Kumankumah, Nigel Boachie Biringa, Chidera Patel, Anjel Shaileshbhai Vesal, Sulaiman Julienne, Dayquan Siska, Charlotte Luján, Manuel Raúl Meléndez Twum-Barimah, Anthony Velazco, Mauricio Chen, Tianwei |
| author_facet | Jaye, Abubakarr Kumankumah, Nigel Boachie Biringa, Chidera Patel, Anjel Shaileshbhai Vesal, Sulaiman Julienne, Dayquan Siska, Charlotte Luján, Manuel Raúl Meléndez Twum-Barimah, Anthony Velazco, Mauricio Chen, Tianwei |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_14229 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | CORPGEN: Simulating Corporate Environments with Autonomous Digital Employees in Multi-Horizon Task Environments Jaye, Abubakarr Kumankumah, Nigel Boachie Biringa, Chidera Patel, Anjel Shaileshbhai Vesal, Sulaiman Julienne, Dayquan Siska, Charlotte Luján, Manuel Raúl Meléndez Twum-Barimah, Anthony Velazco, Mauricio Chen, Tianwei Artificial Intelligence Emerging Technologies Machine Learning 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. |
| title | CORPGEN: Simulating Corporate Environments with Autonomous Digital Employees in Multi-Horizon Task Environments |
| topic | Artificial Intelligence Emerging Technologies Machine Learning |
| url | https://arxiv.org/abs/2602.14229 |