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.
@article{arxiv.2508.15456,
title = {PyTOD: Programmable Task-Oriented Dialogue with Execution Feedback},
author = {Alexandru Coca and Bo-Hsiang Tseng and Pete Boothroyd and Jianpeng Cheng and Mark Gaynor and Zhenxing Zhang and Joe Stacey and Tristan Guigue and Héctor Martinez Alonso and Diarmuid Ó Séaghdha and Anders Johannsen},
journal= {arXiv preprint arXiv:2508.15456},
year = {2025}
}