@inproceedings{lin-etal-2026-reactod,
    title = "{R}eac{TOD}: Bounded Neuro-Symbolic Agentic {NLU} for Zero-Shot Dialogue State Tracking",
    author = "Lin, Yanjun  and
      Xiao, Zimo  and
      Natarajan, Kartik  and
      Sankaranarayanan, Mahesh  and
      Nawanit, Niraj  and
      Parashar, Rakshit  and
      Zhang, Austin  and
      Konaraddi, Karthik  and
      Mote, Rishita  and
      Niu, Wei",
    editor = "Chang, Kai-Wei  and
      Mehrabi, Ninareh  and
      Krishna, Satyapriya  and
      Das, Anubrata  and
      Dhamala, Jwala  and
      Cao, Yang Trista  and
      Kumarage, Tharindu  and
      Ramakrishna, Anil  and
      Christodoulopoulos, Christos  and
      Wan, Yixin  and
      Galystan, Aram  and
      Kumar, Anoop  and
      Gupta, Rahul",
    booktitle = "Proceedings of the 6th Workshop on Trustworthy {NLP} ({T}rust{NLP} 2026)",
    month = jul,
    year = "2026",
    address = "San Diego, California",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-acl-workshops/2026.trustnlp-main.21/",
    pages = "342--352",
    ISBN = "979-8-89176-418-7",
    abstract = "Task-oriented dialogue systems{---}handling transactions, reservations, and service requests{---}require predictable behavior, yet the moderately-sized LLMs needed for practical latency are prone to hallucination and format errors that cascade into incorrect actions (e.g., a hotel booked for the wrong date). We propose ReacTOD, a bounded neuro-symbolic architecture that reformulates NLU as discrete tool calls within a self-correcting ReAct loop governed by deterministic validation. A bounded ReAct loop enables iterative self-correction, improving accuracy by up to 9.3 percentage points over single-pass inference on MultiWOZ. A symbolic validator enforces action compliance, schema conformance, and coreference consistency on every dialogue state update, achieving a 93.1{\%} self-correction rate on intercepted errors and producing structured execution traces. Incremental state prediction and on-demand history retrieval keep prompts compact, empirically improving instruction adherence in parameter-constrained models. On MultiWOZ 2.1, ReacTOD achieves a new zero-shot state-of-the-art: gpt-oss-20B reaches 52.71{\%} joint goal accuracy, surpassing the previous best by 14 percentage points, while Qwen3-8B achieves 47.34{\%} with only 8B parameters. On the Schema-Guided Dialogue (SGD) benchmark, ReacTOD with Claude-Opus-4.6 achieves 80.68{\%} JGA under fully end-to-end evaluation with predicted domains, and Qwen3-32B reaches 64.09{\%}{---}demonstrating cross-benchmark generalization without task-specific training data."
}