AgentCo-op: Retrieval-Based Synthesis of Interoperable Multi-Agent Workflows
The article discusses AgentCo-op, a framework designed for synthesizing interoperable multi-agent workflows. It highlights the challenges of creating workflows in open-ended scientific settings and presents AgentCo-op as a solution that utilizes retrieval-based synthesis. The framework has shown promising results in genomics case studies and various benchmarks, indicating its potential for enhancing automated workflow design.
- ▪AgentCo-op is a retrieval-based synthesis framework for multi-agent workflows.
- ▪It composes reusable skills and tools into executable workflows while applying local repair when failures occur.
- ▪In case studies, AgentCo-op successfully coordinated specialized agents for spatial transcriptomics and gene-set interpretation.
Opening excerpt (first ~120 words) tap to expand
Computer Science > Artificial Intelligence arXiv:2605.20425 (cs) [Submitted on 19 May 2026] Title:AgentCo-op: Retrieval-Based Synthesis of Interoperable Multi-Agent Workflows Authors:Shuaike Shen, Wenduo Cheng, Shike Wang, Mingqian Ma, Jian Ma View a PDF of the paper titled AgentCo-op: Retrieval-Based Synthesis of Interoperable Multi-Agent Workflows, by Shuaike Shen and 4 other authors View PDF HTML (experimental) Abstract:Designing multi-agent workflows is especially difficult in open-ended scientific settings where tasks lack curated training sets, reliable scalar evaluation metrics, and standardized interfaces between existing tools and agents.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at arXiv cs.AI.
Discussion
0 commentsMore from arXiv cs.AI
