# NOEM3A: a Neuro-symbolic Ontology-Enhanced Method for Multi-intent understanding in Mobile Agents

arXiv: https://arxiv.org/abs/2511.19780 · HTML: https://mlsanity.com/paper/2511.19780/

## 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 18 of its 28 references that are in the map.
- First posted on arXiv 24 Nov 2025.

## 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, 18 of 18)

- [Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks](https://mlsanity.com/paper/2005.11401/) (2020)
- [From Local to Global: A Graph RAG Approach to Query-Focused Summarization](https://mlsanity.com/paper/2404.16130/) (2024)
- [MultiWOZ -- A Large-Scale Multi-Domain Wizard-of-Oz Dataset for Task-Oriented Dialogue Modelling](https://mlsanity.com/paper/1810.00278/) (2018)
- [TinyLlama: An Open-Source Small Language Model](https://mlsanity.com/paper/2401.02385/) (2024)
- [ERNIE: Enhanced Language Representation with Informative Entities](https://mlsanity.com/paper/1905.07129/) (2019)
- [MobileBERT: a Compact Task-Agnostic BERT for Resource-Limited Devices](https://mlsanity.com/paper/2004.02984/) (2020)
- [GRAG: Graph Retrieval-Augmented Generation](https://mlsanity.com/paper/2405.16506/) (2024)
- [MultiWOZ 2.3: A multi-domain task-oriented dialogue dataset enhanced with annotation corrections and co-reference annotation](https://mlsanity.com/paper/2010.05594/) (2020)
- [ChatGPT for Zero-shot Dialogue State Tracking: A Solution or an Opportunity?](https://mlsanity.com/paper/2306.01386/) (2023)
- [Large Language Models as Zero-shot Dialogue State Tracker through Function Calling](https://mlsanity.com/paper/2402.10466/) (2024)
- [FolkScope: Intention Knowledge Graph Construction for E-commerce Commonsense Discovery](https://mlsanity.com/paper/2211.08316/) (2022)
- [Generalized Zero-shot Intent Detection via Commonsense Knowledge](https://mlsanity.com/paper/2102.02925/) (2021)
- [MobileQuant: Mobile-friendly Quantization for On-device Language Models](https://mlsanity.com/paper/2408.13933/) (2024)
- [Ontology-Guided Reverse Thinking Makes Large Language Models Stronger on Knowledge Graph Question Answering](https://mlsanity.com/paper/2502.11491/) (2025)
- [PhoneLM:an Efficient and Capable Small Language Model Family through Principled Pre-training](https://mlsanity.com/paper/2411.05046/) (2024)
- [Inference is All You Need: Self Example Retriever for Cross-domain Dialogue State Tracking with ChatGPT](https://mlsanity.com/paper/2409.06243/) (2024)
- [Creating an LLM-based AI-agent: A high-level methodology towards enhancing LLMs with APIs](https://mlsanity.com/paper/2412.13233/) (2024)
- [A Concept Knowledge Graph for User Next Intent Prediction at Alipay](https://mlsanity.com/paper/2301.00503/) (2023)

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

- [A generative-informed neuro-symbolic framework for syntactic ambiguity resolution: Evidence from Arabic DPs](https://mlsanity.com/paper/2610.02529/) (2026)

## Abstract (arXiv)

Mobile agents must map natural-language requests to executable intents under tight latency and privacy constraints. Scaling the language model is often an inefficient way to improve this component. We present NOEM3A, a lightweight neuro-symbolic layer that augments compact language models with an intent ontology. For each query, NOEM3A retrieves a small ontology neighborhood, injects candidate action labels into the prompt and applies a token-level decoding prior toward valid labels. This injects symbolic intent structure into both input and output representations while keeping inference suitable for local interaction. We also use Semantic Intent Similarity (SIS), a hierarchy-aware diagnostic based on ontology depth, to capture semantic proximity when predicted intents differ lexically. Experiments on MultiWOZ 2.3 dialogues show that ontology augmentation consistently improves TinyLlama and Llama-3.2-3B on exact match and Slot-F1. Our results suggest that symbolic alignment is an effective strategy for accurate, responsive and privacy-preserving on-device NLU.