EUVIMEDEUVIMED entdecken
Crossref

FusionAdvisor: Decision guidance for efficient operator fusion in deep models

Chang Jiang, Yanan Yang, Shen Gao, Jianjiang Li, Jie Wu

Big Data Mining and Analytics · 20.11.2026

DOI: 10.26599/bdma.2026.9020025 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Big Data Mining and Analytics, online erschienen am 20.11.2026. Verfasst von Chang Jiang und 4 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

As deep learning models grow in size and complexity, optimizing the execution efficiency becomes critical. Operator fusion merges multiple operators to reduce memory overhead and enhance computational efficiency. However, the combinatorial explosion of possible fusion combinations in modern models creates a search space that challenges traditional manual and heuristic methods. We propose FusionAdvisor, a neural network based framework that predicts the speedups of candidate fusion strategies from computation graphs and recommends the top- candidates (recommended strategies). Trained on our constructed fusion strategy dataset, it abstracts fusion selection as a service, enabling developers to choose high-performance strategies and accelerate inference efficiently. Experimental results demonstrate that FusionAdvisor’s neural network predicts post-fusion speedups with low error, enabling promising strategy recommendations while maintaining search efficiency comparable to the existing methods. This work advances automated operator fusion selection, reduces optimization costs, and improves resource utilization in complex scenarios, including Large Language Models (LLMs) and other large-scale models.

Zitieren

APA

Jiang, C., Yang, Y., Gao, S., Li, J., & Wu, J. (2026). FusionAdvisor: Decision guidance for efficient operator fusion in deep models. Big Data Mining and Analytics. https://doi.org/10.26599/bdma.2026.9020025

Vancouver

Jiang C, Yang Y, Gao S, Li J, Wu J. FusionAdvisor: Decision guidance for efficient operator fusion in deep models. Big Data Mining and Analytics. 2026. doi:10.26599/bdma.2026.9020025

BibTeX
@article{Jiang2026,
  author = {Jiang, Chang and Yang, Yanan and Gao, Shen and Li, Jianjiang and Wu, Jie},
  title = {FusionAdvisor: Decision guidance for efficient operator fusion in deep models},
  journal = {Big Data Mining and Analytics},
  year = {2026},
  doi = {10.26599/bdma.2026.9020025}
}
Kostenlos starten →