Unleashing the Power of Tree-of-Thoughts for Edge-Enabled AIGC Service Provisioning
Zhang Liu, Shanhao Zhan, Shaowei Shen, Lianfen Huang, Qiao Xiang, Ying-Jun Angela Zhang, Dusit Niyato

TL;DR
This paper explores leveraging Tree-of-Thoughts prompting in edge-enabled AIGC services, proposing a DAG model and a diffusion-based RL algorithm to optimize reasoning paths for reduced latency and maintained quality.
Contribution
It introduces a DAG-based model for reasoning in ToT prompting and a diffusion-based RL algorithm for optimal thought assignment in resource-constrained edge environments.
Findings
DSAC reduces total generation delay by up to 36.09% compared to baseline algorithms.
Latency is reduced by over 80% compared to fully local generation under strict quality constraints.
The DAG model accurately characterizes the reasoning process in ToT prompting.
Abstract
Delivering AI-generated content (AIGC) services fundamentally relies on the reasoning capabilities of generative AI (GenAI) models. Chain-of-Thought (CoT) enhances such reasoning by guiding models through intermediate steps, while Tree-of-Thoughts (ToT) further extends CoT by exploring multiple candidate reasoning paths simultaneously, thereby greatly improving AIGC service quality. However, generating diverse reasoning paths requires separate calls to computationally intensive GenAI models, posing significant challenges for resource constrained user devices. In this paper, we investigate mobile edge computing-enabled AIGC service provisioning with ToT prompting. Specifically, using creative writing AIGC tasks as a case study, we first characterize the number of output tokens as a measure of computational resources in GenAI models and establish its relationship with generation delay and…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
