Self-Manager: Parallel Agent Loop for Long-form Deep Research
This addresses scalability and adaptability issues in multi-faceted research tasks for AI agents, though it appears incremental as it builds on existing agent frameworks.
The paper tackles the problem of mutual interference and blocking in long-form deep research tasks by introducing Self-Manager, a parallel agent loop that enables asynchronous and concurrent execution, which consistently outperforms existing single-agent loop baselines across all metrics on DeepResearch Bench.
Long-form deep research requires multi-faceted investigations over extended horizons to get a comprehensive report. When handling such complex tasks, existing agents manage context at the subtask level to overcome linear context accumulation and information loss. However, they still adhere to a single context window and sequential execution paradigm, which results in mutual interference and blocking behavior, restricting scalability and adaptability. To address this issue, this paper introduces Self-Manager, a parallel agent loop that enables asynchronous and concurrent execution. The main thread can create multiple subthreads, each with its own isolated context, and manage them iteratively through Thread Control Blocks, allowing for more focused and flexible parallel agent execution. To assess its effectiveness, we benchmark Self-Manager on DeepResearch Bench, where it consistently outperforms existing single-agent loop baselines across all metrics. Furthermore, we conduct extensive analytical experiments to demonstrate the necessity of Self-Manager's design choices, as well as its advantages in contextual capacity, efficiency, and generalization.