Paper page - Mind DeepResearch Technical Report

Source: https://huggingface.co/papers/2604.14518

Abstract

MindDR is an efficient multi-agent deep research framework that achieves high performance through a collaborative three-agent architecture and specialized four-stage training pipeline, demonstrating strong results on multiple benchmarks.

We present Mind DeepResearch (MindDR), an efficientmulti-agent deep research frameworkthat achieves leading performance with only ~30B-parameter models through a meticulously designeddata synthesisandmulti-stage training pipeline. The core innovation of MindDR lies in acollaborative three-agent architecture(Planning Agent, DeepSearch Agent, and Report Agent) and a four-stageagent-specialized training pipelinecomprisingSFT cold-start,Search-RL,Report-RLandpreference alignment. With this regime, MindDR demonstrates competitive performance even with ~30B-scale models. Specifically, MindDR achieves 45.7% on BrowseComp-ZH, 42.8% on BrowseComp, 46.5% on WideSearch, 75.0% on xbench-DS, and 52.5 on DeepResearch Bench, outperforming comparable-scale open-source agent systems and rivaling larger-scale models. MindDR has been deployed as an online product in Li Auto. Furthermore, we introduce MindDR Bench, a curated benchmark of 500real-world Chinese queriesfrom our internal product user interactions, evaluated through a comprehensivemulti-dimensional rubric systemrather than relying on a single RACE metric. On MindDR Bench, MindDR achieves a state-of-the-art score of 51.8.

View arXiv pageView PDFAdd to collection

Get this paper in your agent:

hf papers read 2604\.14518

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2604.14518 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2604.14518 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2604.14518 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.