MermaidFlow: Redefining Agentic Workflow Generation via Safety-Constrained Evolutionary Programming
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866910973273571328 |
|---|---|
| author | Zheng, Chengqi Chen, Jianda Lyu, Yueming Ng, Wen Zheng Terence Zhang, Haopeng Ong, Yew-Soon Tsang, Ivor Yin, Haiyan |
| author_facet | Zheng, Chengqi Chen, Jianda Lyu, Yueming Ng, Wen Zheng Terence Zhang, Haopeng Ong, Yew-Soon Tsang, Ivor Yin, Haiyan |
| contents | Despite the promise of autonomous agentic reasoning, existing workflow generation methods frequently produce fragile, unexecutable plans due to unconstrained LLM-driven construction. We introduce MermaidFlow, a framework that redefines the agentic search space through safety-constrained graph evolution. At its core, MermaidFlow represent workflows as a verifiable intermediate representation using Mermaid, a structured and human-interpretable graph language. We formulate domain-aware evolutionary operators, i.e., crossover, mutation, insertion, and deletion, to preserve semantic correctness while promoting structural diversity, enabling efficient exploration of a high-quality, statically verifiable workflow space. Without modifying task settings or evaluation protocols, MermaidFlow achieves consistent improvements in success rates and faster convergence to executable plans on the agent reasoning benchmark. The experimental results demonstrate that safety-constrained graph evolution offers a scalable, modular foundation for robust and interpretable agentic reasoning systems. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_22967 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | MermaidFlow: Redefining Agentic Workflow Generation via Safety-Constrained Evolutionary Programming Zheng, Chengqi Chen, Jianda Lyu, Yueming Ng, Wen Zheng Terence Zhang, Haopeng Ong, Yew-Soon Tsang, Ivor Yin, Haiyan Machine Learning Multiagent Systems Despite the promise of autonomous agentic reasoning, existing workflow generation methods frequently produce fragile, unexecutable plans due to unconstrained LLM-driven construction. We introduce MermaidFlow, a framework that redefines the agentic search space through safety-constrained graph evolution. At its core, MermaidFlow represent workflows as a verifiable intermediate representation using Mermaid, a structured and human-interpretable graph language. We formulate domain-aware evolutionary operators, i.e., crossover, mutation, insertion, and deletion, to preserve semantic correctness while promoting structural diversity, enabling efficient exploration of a high-quality, statically verifiable workflow space. Without modifying task settings or evaluation protocols, MermaidFlow achieves consistent improvements in success rates and faster convergence to executable plans on the agent reasoning benchmark. The experimental results demonstrate that safety-constrained graph evolution offers a scalable, modular foundation for robust and interpretable agentic reasoning systems. |
| title | MermaidFlow: Redefining Agentic Workflow Generation via Safety-Constrained Evolutionary Programming |
| topic | Machine Learning Multiagent Systems |
| url | https://arxiv.org/abs/2505.22967 |