DR-Venus: Towards Frontier Edge-Scale Deep Research Agents with Only 10K Open Data

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Venus Team, Dai, Sunhao, Deng, Yong, Lin, Jinzhen, Song, Yusheng, Wang, Guoqing, Wu, Xiaofeng, Zhou, Yuqi, Yang, Shuo, Ying, Zhenzhe, Zhang, Zhanwei, Meng, Changhua, Wang, Weiqiang
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914497651802112
author Venus Team
Dai, Sunhao
Deng, Yong
Lin, Jinzhen
Song, Yusheng
Wang, Guoqing
Wu, Xiaofeng
Zhou, Yuqi
Yang, Shuo
Ying, Zhenzhe
Zhang, Zhanwei
Meng, Changhua
Wang, Weiqiang
author_facet Venus Team
Dai, Sunhao
Deng, Yong
Lin, Jinzhen
Song, Yusheng
Wang, Guoqing
Wu, Xiaofeng
Zhou, Yuqi
Yang, Shuo
Ying, Zhenzhe
Zhang, Zhanwei
Meng, Changhua
Wang, Weiqiang
contents Edge-scale deep research agents based on small language models are attractive for real-world deployment due to their advantages in cost, latency, and privacy. In this work, we study how to train a strong small deep research agent under limited open-data by improving both data quality and data utilization. We present DR-Venus, a frontier 4B deep research agent for edge-scale deployment, built entirely on open data. Our training recipe consists of two stages. In the first stage, we use agentic supervised fine-tuning (SFT) to establish basic agentic capability, combining strict data cleaning with resampling of long-horizon trajectories to improve data quality and utilization. In the second stage, we apply agentic reinforcement learning (RL) to further improve execution reliability on long-horizon deep research tasks. To make RL effective for small agents in this setting, we build on IGPO and design turn-level rewards based on information gain and format-aware regularization, thereby enhancing supervision density and turn-level credit assignment. Built entirely on roughly 10K open-data, DR-Venus-4B significantly outperforms prior agentic models under 9B parameters on multiple deep research benchmarks, while also narrowing the gap to much larger 30B-class systems. Our further analysis shows that 4B agents already possess surprisingly strong performance potential, highlighting both the deployment promise of small models and the value of test-time scaling in this setting. We release our models, code, and key recipes to support reproducible research on edge-scale deep research agents.
format Preprint
id arxiv_https___arxiv_org_abs_2604_19859
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle DR-Venus: Towards Frontier Edge-Scale Deep Research Agents with Only 10K Open Data
Venus Team
Dai, Sunhao
Deng, Yong
Lin, Jinzhen
Song, Yusheng
Wang, Guoqing
Wu, Xiaofeng
Zhou, Yuqi
Yang, Shuo
Ying, Zhenzhe
Zhang, Zhanwei
Meng, Changhua
Wang, Weiqiang
Machine Learning
Artificial Intelligence
Computation and Language
Information Retrieval
Edge-scale deep research agents based on small language models are attractive for real-world deployment due to their advantages in cost, latency, and privacy. In this work, we study how to train a strong small deep research agent under limited open-data by improving both data quality and data utilization. We present DR-Venus, a frontier 4B deep research agent for edge-scale deployment, built entirely on open data. Our training recipe consists of two stages. In the first stage, we use agentic supervised fine-tuning (SFT) to establish basic agentic capability, combining strict data cleaning with resampling of long-horizon trajectories to improve data quality and utilization. In the second stage, we apply agentic reinforcement learning (RL) to further improve execution reliability on long-horizon deep research tasks. To make RL effective for small agents in this setting, we build on IGPO and design turn-level rewards based on information gain and format-aware regularization, thereby enhancing supervision density and turn-level credit assignment. Built entirely on roughly 10K open-data, DR-Venus-4B significantly outperforms prior agentic models under 9B parameters on multiple deep research benchmarks, while also narrowing the gap to much larger 30B-class systems. Our further analysis shows that 4B agents already possess surprisingly strong performance potential, highlighting both the deployment promise of small models and the value of test-time scaling in this setting. We release our models, code, and key recipes to support reproducible research on edge-scale deep research agents.
title DR-Venus: Towards Frontier Edge-Scale Deep Research Agents with Only 10K Open Data
topic Machine Learning
Artificial Intelligence
Computation and Language
Information Retrieval
url https://arxiv.org/abs/2604.19859