DyFlow: Dynamic Workflow Framework for Agentic Reasoning

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Wang, Yanbo, Xu, Zixiang, Huang, Yue, Wang, Xiangqi, Song, Zirui, Gao, Lang, Wang, Chenxi, Tang, Xiangru, Zhao, Yue, Cohan, Arman, Zhang, Xiangliang, Chen, Xiuying
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866916979803160576
author Wang, Yanbo
Xu, Zixiang
Huang, Yue
Wang, Xiangqi
Song, Zirui
Gao, Lang
Wang, Chenxi
Tang, Xiangru
Zhao, Yue
Cohan, Arman
Zhang, Xiangliang
Chen, Xiuying
author_facet Wang, Yanbo
Xu, Zixiang
Huang, Yue
Wang, Xiangqi
Song, Zirui
Gao, Lang
Wang, Chenxi
Tang, Xiangru
Zhao, Yue
Cohan, Arman
Zhang, Xiangliang
Chen, Xiuying
contents Agent systems based on large language models (LLMs) have shown great potential in complex reasoning tasks, but building efficient and generalizable workflows remains a major challenge. Most existing approaches rely on manually designed processes, which limits their adaptability across different tasks. While a few methods attempt automated workflow generation, they are often tied to specific datasets or query types and make limited use of intermediate feedback, reducing system robustness and reasoning depth. Moreover, their operations are typically predefined and inflexible. To address these limitations, we propose DyFlow, a dynamic workflow generation framework that adaptively constructs and adjusts reasoning procedures based on task requirements and real-time intermediate feedback, thereby enhancing cross-task generalization. DyFlow consists of two core components: a designer and an executor. The designer decomposes complex problems into a sequence of sub-goals defined by high-level objectives and dynamically plans the next steps based on intermediate outputs and feedback. These plans are then carried out by the executor, which executes each operation using dynamic operators with context-aware parameterization, enabling flexible and semantically grounded reasoning. We systematically evaluate DyFlow across diverse domains, including social reasoning, biomedical tasks, mathematical problem solving, and code generation. Results demonstrate that DyFlow significantly outperforms existing baselines, achieving substantial Pass@k improvements and exhibiting robust generalization across diverse domains. The code is publicly available at https://github.com/wyf23187/DyFlow.
format Preprint
id arxiv_https___arxiv_org_abs_2509_26062
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle DyFlow: Dynamic Workflow Framework for Agentic Reasoning
Wang, Yanbo
Xu, Zixiang
Huang, Yue
Wang, Xiangqi
Song, Zirui
Gao, Lang
Wang, Chenxi
Tang, Xiangru
Zhao, Yue
Cohan, Arman
Zhang, Xiangliang
Chen, Xiuying
Computation and Language
Agent systems based on large language models (LLMs) have shown great potential in complex reasoning tasks, but building efficient and generalizable workflows remains a major challenge. Most existing approaches rely on manually designed processes, which limits their adaptability across different tasks. While a few methods attempt automated workflow generation, they are often tied to specific datasets or query types and make limited use of intermediate feedback, reducing system robustness and reasoning depth. Moreover, their operations are typically predefined and inflexible. To address these limitations, we propose DyFlow, a dynamic workflow generation framework that adaptively constructs and adjusts reasoning procedures based on task requirements and real-time intermediate feedback, thereby enhancing cross-task generalization. DyFlow consists of two core components: a designer and an executor. The designer decomposes complex problems into a sequence of sub-goals defined by high-level objectives and dynamically plans the next steps based on intermediate outputs and feedback. These plans are then carried out by the executor, which executes each operation using dynamic operators with context-aware parameterization, enabling flexible and semantically grounded reasoning. We systematically evaluate DyFlow across diverse domains, including social reasoning, biomedical tasks, mathematical problem solving, and code generation. Results demonstrate that DyFlow significantly outperforms existing baselines, achieving substantial Pass@k improvements and exhibiting robust generalization across diverse domains. The code is publicly available at https://github.com/wyf23187/DyFlow.
title DyFlow: Dynamic Workflow Framework for Agentic Reasoning
topic Computation and Language
url https://arxiv.org/abs/2509.26062