Don\'t Stop Early: Scalable Enterprise Deep Research with Controlled Information Flow and Evidence-Aware Termination

TL;DR

This paper introduces a scalable enterprise deep research architecture that improves decision-ready report generation by controlling information flow, localizing context, and enforcing evidence-based termination criteria.

Contribution

It proposes a novel system design that decomposes research requests, localizes context, and enforces evidence-based stopping to enhance enterprise research quality.

Findings

Achieves the strongest performance on internal and public benchmarks.

Reduces premature stopping and improves report depth.

Enhances consistency and comprehensiveness of research outputs.

Abstract

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.