Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

InternAgent-1.5: A Unified Agentic Framework for Long-Horizon Autonomous Scientific Discovery

Shiyang Feng, Runmin Ma, Xiangchao Yan, Yue Fan, Yusong Hu, Songtao Huang, Shuaiyu Zhang, Zongsheng Cao, Tianshuo Peng, Jiakang Yuan, Zijie Guo, Zhijie Zhong, Shangheng Du, Weida Wang, Jinxin Shi, Yuhao Zhou, Xiaohan He, Zhiyin Yu, Fangchen Yu, Qihao Zheng, Jiamin Wu, Mianxin Liu, Chi Zhang, Shaowei Hou, Shuya Li, Yankai Jiang, Wenjie Lou, Lilong Wang, Zifu Wang, Jiong Wang, Wanghan Xu, Yue Deng, Dongrui Liu, Yiheng Wang, Wenlong Zhang, Fenghua Ling, Shufei Zhang, Xiaosong Wang, Shuangjia Zheng, Xun Huang, Siqi Sun, Shuyue Hu, Peng Ye, Chunfeng Song, Bin Wang, Conghui He, Yihao Liu, Xin Li, Qibin Hou, Tao Chen, Xiangyu Yue, Bin Wang, Liang He, Dahua Lin, Bowen Zhou, Bo Zhang, Lei Bai

TL;DR

InternAgent-1.5 introduces a unified, three-subsystem framework (Generation, Verification, Evolution) for end-to-end, long-horizon autonomous scientific discovery across computational and empirical domains. It leverages a Cross-Disciplinary Knowledge Graph and a Flow Graph to enable dynamic planning, multi-domain reasoning, and graph-guided output synthesis, supported by Structured Cognitive Memory (Strategy-Procedural, Task-Episodic, Semantic-Knowledge) for sustained improvement. The approach achieves leading performance on standard agentic reasoning benchmarks (GAIA, HLE, GPQA, FrontierScience) and demonstrates robust capabilities in algorithm discovery and empirical discovery across Earth, Life, Biological, and Physical sciences, including automated climate analysis, target identification, and reaction outcome prediction. This work demonstrates a general, scalable architecture for cross-disciplinary autonomous scientific discovery with demonstrated potential for continuous improvement and real-world impact in scientific workflows.

Abstract

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.

InternAgent-1.5: A Unified Agentic Framework for Long-Horizon Autonomous Scientific Discovery

TL;DR

InternAgent-1.5 introduces a unified, three-subsystem framework (Generation, Verification, Evolution) for end-to-end, long-horizon autonomous scientific discovery across computational and empirical domains. It leverages a Cross-Disciplinary Knowledge Graph and a Flow Graph to enable dynamic planning, multi-domain reasoning, and graph-guided output synthesis, supported by Structured Cognitive Memory (Strategy-Procedural, Task-Episodic, Semantic-Knowledge) for sustained improvement. The approach achieves leading performance on standard agentic reasoning benchmarks (GAIA, HLE, GPQA, FrontierScience) and demonstrates robust capabilities in algorithm discovery and empirical discovery across Earth, Life, Biological, and Physical sciences, including automated climate analysis, target identification, and reaction outcome prediction. This work demonstrates a general, scalable architecture for cross-disciplinary autonomous scientific discovery with demonstrated potential for continuous improvement and real-world impact in scientific workflows.

Abstract

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.
Paper Structure (47 sections, 9 equations, 42 figures, 11 tables)

This paper contains 47 sections, 9 equations, 42 figures, 11 tables.

Table of Contents

  1. Introduction
  2. InternAgent-1.5
  3. System Overview
  4. Architecture
  5. Foundational Capabilities
  6. Cross Disciplinary Graph Construction and Knowledge Capturing
  7. Cross-Disciplinary Knowledge Graph
  8. Graph Construction
  9. Knowledge Extraction and Retrieval
  10. Flow Graph
  11. Structured Knowledge Flow
  12. Dynamic planning and refinement
  13. Graph-Guided Output Synthesis
  14. Cross-Disciplinary Knowledge Collector.
  15. Reasoning capability enhancement
  16. ...and 32 more sections

Figures (42)

  • Figure 1: Performance comparison of InternAgent-1.5 across GAIA mialon2023gaia, GPQA rein2024gpqa, HLE-full phan2025humanity, and FrontierScience frontierscience.
  • Figure 2: Overview of InternAgent-1.5 that summarizes its foundational capabilities, unified discovery pipeline, and supported scientific tasks in a high‑level manner.
  • Figure 3: Overview of InternAgent-1.5, illustrating its unified scientific discovery pipeline organized around the Generation, Verification, and Evolution subsystems. The system operates through foundational capabilities for deep research, solution refinement, and long horizon memory, which together enable sustained autonomous scientific discovery.
  • Figure 4: The illustration for our cross-disciplinary knowledge graph.
  • Figure 5: The illustration for our flow graph.
  • ...and 37 more figures