Interactive Speculative Planning: Enhance Agent Efficiency through Co-design of System and User Interface
Wenyue Hua, Mengting Wan, Shashank Vadrevu, Ryan Nadel, Yongfeng, Zhang, and Chi Wang

TL;DR
This paper introduces Interactive Speculative Planning, a human-centered approach that co-designs system and interface to improve the efficiency of large language model agents by leveraging human interactions and interruptions.
Contribution
It proposes a novel co-designed system and interface for LLM agents that incorporates human interruptions to reduce planning latency and enhance user-centric efficiency.
Findings
Improved planning efficiency through human-in-the-loop interactions
System design that manages user interruptions effectively
Enhanced agent responsiveness and user satisfaction
Abstract
Agents, as user-centric tools, are increasingly deployed for human task delegation, assisting with a broad spectrum of requests by generating thoughts, engaging with user proxies, and producing action plans. However, agents based on large language models (LLMs) often face substantial planning latency due to two primary factors: the efficiency limitations of the underlying LLMs due to their large size and high demand, and the structural complexity of the agents due to the extensive generation of intermediate thoughts to produce the final output. Given that inefficiency in service provision can undermine the value of automation for users, this paper presents a human-centered efficient agent planning method -- Interactive Speculative Planning -- aiming at enhancing the efficiency of agent planning through both system design and human-AI interaction. Our approach advocates for the co-design…
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.
Taxonomy
TopicsComplex Systems and Decision Making · Multi-Agent Systems and Negotiation · Transportation and Mobility Innovations
Methodstravel james
