🤖 AI Summary
This work addresses the challenge of generating high-quality, comprehensive, and reliable business reports from vast amounts of noisy web data to support high-stakes decision-making. The authors propose a training-free agentic workflow that emulates the cognitive process of professional business analysts. Their approach integrates fine-grained intent parsing, dynamic web retrieval, real-time information distillation, and iterative report generation, augmented by a dynamic memory mechanism to enhance the long-horizon reasoning capabilities of large language models. Evaluated on QRC-Eval—a newly curated benchmark comprising 200 real-world business tasks—the proposed method significantly outperforms state-of-the-art deep research agents from OpenAI and Gemini, achieving expert-level performance in automated business report synthesis.
📝 Abstract
Synthesizing informative commercial reports from massive and noisy web sources is critical for high-stakes business decisions. Although current deep research agents achieve notable progress, their reports still remain limited in terms of quality, reliability, and coverage. In this work, we propose Mind2Report, a cognitive deep research agent that emulates the commercial analyst to synthesize expert-level reports. Specifically, it first probes fine-grained intent, then searches web sources and records distilled information on the fly, and subsequently iteratively synthesizes the report. We design Mind2Report as a training-free agentic workflow that augments general large language models (LLMs) with dynamic memory to support these long-form cognitive processes. To rigorously evaluate Mind2Report, we further construct QRC-Eval comprising 200 real-world commercial tasks and establish a holistic evaluation strategy to assess report quality, reliability, and coverage. Experiments demonstrate that Mind2Report outperforms leading baselines, including OpenAI and Gemini deep research agents. Although this is a preliminary study, we expect it to serve as a foundation for advancing the future design of commercial deep research agents. Our code and data are available at https://github.com/Melmaphother/Mind2Report.