_version_ 1866918328881119232
author Feng, Shiyang
Ma, Runmin
Yan, Xiangchao
Fan, Yue
Hu, Yusong
Huang, Songtao
Zhang, Shuaiyu
Cao, Zongsheng
Peng, Tianshuo
Yuan, Jiakang
Guo, Zijie
Zhong, Zhijie
Du, Shangheng
Wang, Weida
Shi, Jinxin
Zhou, Yuhao
He, Xiaohan
Yu, Zhiyin
Yu, Fangchen
Zheng, Qihao
Wu, Jiamin
Liu, Mianxin
Zhang, Chi
Hou, Shaowei
Li, Shuya
Jiang, Yankai
Lou, Wenjie
Wang, Lilong
Wang, Zifu
Wang, Jiong
Xu, Wanghan
Deng, Yue
Liu, Dongrui
Wang, Yiheng
Zhang, Wenlong
Ling, Fenghua
Zhang, Shufei
Wang, Xiaosong
Zheng, Shuangjia
Huang, Xun
Sun, Siqi
Hu, Shuyue
Ye, Peng
Song, Chunfeng
Wang, Bin
He, Conghui
Liu, Yihao
Li, Xin
Hou, Qibin
Chen, Tao
Yue, Xiangyu
Wang, Bin
He, Liang
Lin, Dahua
Zhou, Bowen
Zhang, Bo
Bai, Lei
author_facet Feng, Shiyang
Ma, Runmin
Yan, Xiangchao
Fan, Yue
Hu, Yusong
Huang, Songtao
Zhang, Shuaiyu
Cao, Zongsheng
Peng, Tianshuo
Yuan, Jiakang
Guo, Zijie
Zhong, Zhijie
Du, Shangheng
Wang, Weida
Shi, Jinxin
Zhou, Yuhao
He, Xiaohan
Yu, Zhiyin
Yu, Fangchen
Zheng, Qihao
Wu, Jiamin
Liu, Mianxin
Zhang, Chi
Hou, Shaowei
Li, Shuya
Jiang, Yankai
Lou, Wenjie
Wang, Lilong
Wang, Zifu
Wang, Jiong
Xu, Wanghan
Deng, Yue
Liu, Dongrui
Wang, Yiheng
Zhang, Wenlong
Ling, Fenghua
Zhang, Shufei
Wang, Xiaosong
Zheng, Shuangjia
Huang, Xun
Sun, Siqi
Hu, Shuyue
Ye, Peng
Song, Chunfeng
Wang, Bin
He, Conghui
Liu, Yihao
Li, Xin
Hou, Qibin
Chen, Tao
Yue, Xiangyu
Wang, Bin
He, Liang
Lin, Dahua
Zhou, Bowen
Zhang, Bo
Bai, Lei
contents We introduce InternAgent-1.5, a unified system designed for end-to-end scientific discovery across computational and empirical domains. The system is built on a structured architecture composed of three coordinated subsystems for generation, verification, and evolution. These subsystems are supported by foundational capabilities for deep research, solution optimization, and long horizon memory. The architecture allows InternAgent-1.5 to operate continuously across extended discovery cycles while maintaining coherent and improving behavior. It also enables the system to coordinate computational modeling and laboratory experimentation within a single unified system. We evaluate InternAgent-1.5 on scientific reasoning benchmarks such as GAIA, HLE, GPQA, and FrontierScience, and the system achieves leading performance that demonstrates strong foundational capabilities. Beyond these benchmarks, we further assess two categories of discovery tasks. In algorithm discovery tasks, InternAgent-1.5 autonomously designs competitive methods for core machine learning problems. In empirical discovery tasks, it executes complete computational or wet lab experiments and produces scientific findings in earth, life, biological, and physical domains. Overall, these results show that InternAgent-1.5 provides a general and scalable framework for autonomous scientific discovery.
format Preprint
id arxiv_https___arxiv_org_abs_2602_08990
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle InternAgent-1.5: A Unified Agentic Framework for Long-Horizon Autonomous Scientific Discovery
Feng, Shiyang
Ma, Runmin
Yan, Xiangchao
Fan, Yue
Hu, Yusong
Huang, Songtao
Zhang, Shuaiyu
Cao, Zongsheng
Peng, Tianshuo
Yuan, Jiakang
Guo, Zijie
Zhong, Zhijie
Du, Shangheng
Wang, Weida
Shi, Jinxin
Zhou, Yuhao
He, Xiaohan
Yu, Zhiyin
Yu, Fangchen
Zheng, Qihao
Wu, Jiamin
Liu, Mianxin
Zhang, Chi
Hou, Shaowei
Li, Shuya
Jiang, Yankai
Lou, Wenjie
Wang, Lilong
Wang, Zifu
Wang, Jiong
Xu, Wanghan
Deng, Yue
Liu, Dongrui
Wang, Yiheng
Zhang, Wenlong
Ling, Fenghua
Zhang, Shufei
Wang, Xiaosong
Zheng, Shuangjia
Huang, Xun
Sun, Siqi
Hu, Shuyue
Ye, Peng
Song, Chunfeng
Wang, Bin
He, Conghui
Liu, Yihao
Li, Xin
Hou, Qibin
Chen, Tao
Yue, Xiangyu
Wang, Bin
He, Liang
Lin, Dahua
Zhou, Bowen
Zhang, Bo
Bai, Lei
Artificial Intelligence
We introduce InternAgent-1.5, a unified system designed for end-to-end scientific discovery across computational and empirical domains. The system is built on a structured architecture composed of three coordinated subsystems for generation, verification, and evolution. These subsystems are supported by foundational capabilities for deep research, solution optimization, and long horizon memory. The architecture allows InternAgent-1.5 to operate continuously across extended discovery cycles while maintaining coherent and improving behavior. It also enables the system to coordinate computational modeling and laboratory experimentation within a single unified system. We evaluate InternAgent-1.5 on scientific reasoning benchmarks such as GAIA, HLE, GPQA, and FrontierScience, and the system achieves leading performance that demonstrates strong foundational capabilities. Beyond these benchmarks, we further assess two categories of discovery tasks. In algorithm discovery tasks, InternAgent-1.5 autonomously designs competitive methods for core machine learning problems. In empirical discovery tasks, it executes complete computational or wet lab experiments and produces scientific findings in earth, life, biological, and physical domains. Overall, these results show that InternAgent-1.5 provides a general and scalable framework for autonomous scientific discovery.
title InternAgent-1.5: A Unified Agentic Framework for Long-Horizon Autonomous Scientific Discovery
topic Artificial Intelligence
url https://arxiv.org/abs/2602.08990