# ReacTOD: Bounded Neuro-Symbolic Agentic NLU for Zero-Shot Dialogue State Tracking

arXiv: https://arxiv.org/abs/2605.19077 · HTML: https://mlsanity.com/paper/2605.19077/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in General LLM agent frameworks and benchmarks, one of 16 directions in LLM agents and tool use.
- Builds on 16 of its 23 references that are in the map.
- First posted on arXiv 18 May 2026.

## Placed in

- LLM agents and tool use / [General LLM agent frameworks and benchmarks](https://mlsanity.com/agents/general-llm-agent-frameworks-and-benchmarks/)

## Builds on (most cited first, 16 of 16)

- [ReAct: Synergizing Reasoning and Acting in Language Models](https://mlsanity.com/paper/2210.03629/) (2022)
- [Self-Refine: Iterative Refinement with Self-Feedback](https://mlsanity.com/paper/2303.17651/) (2023)
- [gpt-oss-120b & gpt-oss-20b Model Card](https://mlsanity.com/paper/2508.10925/) (2025)
- [MultiWOZ -- A Large-Scale Multi-Domain Wizard-of-Oz Dataset for Task-Oriented Dialogue Modelling](https://mlsanity.com/paper/1810.00278/) (2018)
- [Towards Scalable Multi-domain Conversational Agents: The Schema-Guided Dialogue Dataset](https://mlsanity.com/paper/1909.05855/) (2019)
- [Transferable Multi-Domain State Generator for Task-Oriented Dialogue Systems](https://mlsanity.com/paper/1905.08743/) (2019)
- [MultiWOZ 2.2 : A Dialogue Dataset with Additional Annotation Corrections and State Tracking Baselines](https://mlsanity.com/paper/2007.12720/) (2020)
- [BERT for Joint Intent Classification and Slot Filling](https://mlsanity.com/paper/1902.10909/) (2019)
- [MultiWOZ 2.4: A Multi-Domain Task-Oriented Dialogue Dataset with Essential Annotation Corrections to Improve State Tracking Evaluation](https://mlsanity.com/paper/2104.00773/) (2021)
- [Description-Driven Task-Oriented Dialog Modeling](https://mlsanity.com/paper/2201.08904/) (2022)
- [Large Language Models as Zero-shot Dialogue State Tracker through Function Calling](https://mlsanity.com/paper/2402.10466/) (2024)
- [Evaluating Small Language Models for News Summarization: Implications and Factors Influencing Performance](https://mlsanity.com/paper/2502.00641/) (2025)
- [A Fast Attention Network for Joint Intent Detection and Slot Filling on Edge Devices](https://mlsanity.com/paper/2205.07646/) (2022)
- [Inference is All You Need: Self Example Retriever for Cross-domain Dialogue State Tracking with ChatGPT](https://mlsanity.com/paper/2409.06243/) (2024)
- [Exploring ReAct Prompting for Task-Oriented Dialogue: Insights and Shortcomings](https://mlsanity.com/paper/2412.01262/) (2024)
- [A Zero-Shot Open-Vocabulary Pipeline for Dialogue Understanding](https://mlsanity.com/paper/2409.15861/) (2024)

## Built on by (most cited first, 1 of 1)

- [An Auditable Symbolic-RAG-Generative AI Architecture for Goal-Oriented Conversation Orchestration](https://mlsanity.com/paper/2609.07152/) (2026)

## Abstract (arXiv)

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.