arXivDaily arXiv每日学术速递 周一至周五更新
​通知:联邦节假日调休:哥伦布日 / 原住民日(Columbus Day / Indigenous Peoples' Day),今日(13日)无更新。
arXiv 2508.15456cs.CL

PyTOD:具有执行反馈的可编程任务导向对话

PyTOD: Programmable Task-Oriented Dialogue with Execution Feedback

University of Cambridge(剑桥大学);Apple(苹果公司);Meta;Imperial College London(帝国理工学院)
Alexandru Coca, Bo-Hsiang Tseng, Pete Boothroyd, Jianpeng Cheng, Mark Gaynor, Zhenxing Zhang, Joe Stacey, Tristan Guigue, Héctor Martinez Alonso, Diarmuid Ó Séa… 展开作者

Alexandru Coca, Bo-Hsiang Tseng, Pete Boothroyd, Jianpeng Cheng, Mark Gaynor, Zhenxing Zhang, Joe Stacey, Tristan Guigue, Héctor Martinez Alonso, Diarmuid Ó Séaghdha, Anders Johannsen

AI总结:

PyTOD 通过生成可执行代码跟踪对话状态,并利用策略与执行反馈纠错,在 SGD 基准上取得最先进的状态跟踪性能,显著提升任务导向对话的准确性与鲁棒性。

摘要:

可编程任务导向对话(TOD)智能体使语言模型能够遵循结构化对话策略,但其有效性取决于准确的状态跟踪。我们提出 PyTOD,一个生成可执行代码来跟踪对话状态,并利用策略和执行反馈进行高效错误纠正的智能体。为此,PyTOD 采用了一种简单的约束解码方法,使用语言模型而非语法规则来遵循 API 模式。这在具有挑战性的 SGD 基准上实现了最先进的状态跟踪性能。我们的实验表明,随着对话的进行,PyTOD 在准确性和鲁棒的用户目标估计方面都超过了强基线,证明了执行感知状态跟踪的有效性。

英文摘要:

Programmable task-oriented dialogue (TOD) agents enable language models to follow structured dialogue policies, but their effectiveness hinges on accurate state tracking. We present PyTOD, an agent that generates executable code to track dialogue state and uses policy and execution feedback for efficient error correction. To this end, PyTOD employs a simple constrained decoding approach, using a language model instead of grammar rules to follow API schemata. This leads to state-of-the-art state tracking performance on the challenging SGD benchmark. Our experiments show that PyTOD surpasses strong baselines in both accuracy and robust user goal estimation as the dialogue progresses, demonstrating the effectiveness of execution-aware state tracking.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑