![[2508.12685] ToolACE-MT: Non-Autoregressive Generation for Agentic Multi-Turn Interaction](https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png){kind=logo}
[2508.12685] ToolACE-MT: Non-Autoregressive Generation for Agentic Multi-Turn Interaction
Summary
ToolACE-MT introduces a non-autoregressive framework for generating high-quality multi-turn dialogues in agentic interactions, enhancing efficiency and effectiveness in LLM applications.
Why It Matters
As large language models (LLMs) become integral in agentic task-solving, the efficiency of dialogue generation is crucial. ToolACE-MT addresses the limitations of existing autoregressive methods, offering a new paradigm that balances quality and speed, which is essential for real-world applications in AI-driven interactions.
Key Takeaways
- ToolACE-MT utilizes a three-stage process for dialogue generation: initialization, refinement, and verification.
- The framework significantly improves the efficiency of generating multi-turn interactions compared to traditional autoregressive methods.
- Experiments show that ToolACE-MT produces high-quality data suitable for tool-augmented LLM scenarios.
Computer Science > Computation and Language arXiv:2508.12685 (cs) [Submitted on 18 Aug 2025 (v1), last revised 13 Feb 2026 (this version, v3)] Title:ToolACE-MT: Non-Autoregressive Generation for Agentic Multi-Turn Interaction Authors:Xingshan Zeng, Weiwen Liu, Lingzhi Wang, Liangyou Li, Fei Mi, Yasheng Wang, Lifeng Shang, Xin Jiang, Qun Liu View a PDF of the paper titled ToolACE-MT: Non-Autoregressive Generation for Agentic Multi-Turn Interaction, by Xingshan Zeng and 8 other authors View PDF HTML (experimental) Abstract:Agentic task-solving with Large Language Models (LLMs) requires multi-turn, multi-step interactions, often involving complex function calls and dynamic user-agent exchanges. Existing simulation-based data generation methods for such scenarios rely heavily on costly autoregressive interactions between multiple LLM agents, thereby compromising the practical efficiency of agentic data generation. In this paper, we propose ToolACE-MT, a novel Non-Autoregressive Iterative Generation framework for constructing high-quality multi-turn agentic dialogues. ToolACE-MT generates full conversational trajectories through three stages: coarse-grained initialization, iterative refinement, and offline verification. The initialization phase builds a structurally complete yet semantically coarse dialogue skeleton; the iterative refinement phase introduces realistic complexities and continued refinement via mask-and-fill operations; and the offline verification phase ensures ...
