Dont Stop Early: Scalable Enterprise Deep Research with Controlled Information Flow and Evidence-Aware Termination
For enterprises needing decision-ready research reports, this work addresses the practical problems of uneven coverage and premature stopping in deep research systems.
The paper proposes a scalable Enterprise Deep Research architecture that uses outline generation with reflection, dependency-guided context localization, and evidence-based termination criteria to reduce premature stopping and improve report consistency. On the DeepResearch Bench benchmark, the system achieves the strongest overall performance compared to competitive baselines.
Enterprise deep research often fails to produce decision-ready reports due to uneven information coverage, context explosion, and premature stopping. We propose a scalable Enterprise Deep Research (EDR) architecture to address these failures. Our system (i) decomposes requests into coverage-driven objectives via outline generation with reflection, (ii) localizes context with dependency-guided execution and explicit information sharing, and (iii) enforces evidence-based completion criteria so agents iteratively collect information until sufficiency conditions are met. We evaluate on an internal sales enablement task and the public DeepResearch Bench benchmark, where our proposed system design achieves the strongest overall performance compared with competitive deep-research baselines. The results show that dependency-controlled context and explicit evidence sufficiency criteria reduce premature stopping and improve the consistency and depth of enterprise research outputs.