AutoResearch：面向可靠科研工作流自动化的执行接地多智能体框架

View PDF HTML (experimental)

Abstract:Automated research agents increasingly generate code, retrieve literature, and draft scientific artifacts, but they often fail to verify whether generated experiments execute correctly or whether cited sources support generated claims. We present AutoResearch, an execution-grounded multi-agent framework for reliable research workflow automation. AutoResearch couples sandboxed Python/PyTorch execution, iterative code repair, citation verification, claim-support auditing, decision control, and structured \LaTeX{} artifact generation. The system treats runtime errors, citation-verification failures, and review-agent feedback as practical filtering signals for generated research artifacts. In controlled evaluations on HumanEval, MBPP, a SciCode subset, citation-validation tasks, claim-support auditing, and small end-to-end workflow stress tests, AutoResearch improves execution success, citation validity, local claim support, and workflow completion relative to directly comparable baselines. Code-oriented agents are reported separately as partial comparisons. AutoResearch is intended as a reliability-oriented research assistant, not as a fully autonomous scientist or a standalone manuscript-quality benchmark. Source Code: this https URL

Subjects:	Computers and Society (cs.CY); Artificial Intelligence (cs.AI); Software Engineering (cs.SE)
Cite as:	arXiv:2607.02520 [cs.CY]
	(or arXiv:2607.02520v2 [cs.CY] for this version)
	https://doi.org/10.48550/arXiv.2607.02520 arXiv-issued DOI via DataCite

Submission history

From: Rajesh Kumar Raja [view email]
[v1] Mon, 4 May 2026 08:24:21 UTC (4,387 KB)
[v2] Mon, 12 Oct 2026 08:34:34 UTC (847 KB)