# Zero and Few-Shot Localization of Task-Oriented Dialogue Agents with a Distilled Representation

arXiv: https://arxiv.org/abs/2302.09424 · HTML: https://mlsanity.com/paper/2302.09424/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 31 of its 46 references that are in the map.
- First posted on arXiv 18 Feb 2023.

## Builds on (most cited first, 30 of 31)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [HuggingFace's Transformers: State-of-the-art Natural Language Processing](https://mlsanity.com/paper/1910.03771/) (2019)
- [SentencePiece: A simple and language independent subword tokenizer and detokenizer for Neural Text Processing](https://mlsanity.com/paper/1808.06226/) (2018)
- [mT5: A massively multilingual pre-trained text-to-text transformer](https://mlsanity.com/paper/2010.11934/) (2020)
- [Multilingual Denoising Pre-training for Neural Machine Translation](https://mlsanity.com/paper/2001.08210/) (2020)
- [MultiWOZ -- A Large-Scale Multi-Domain Wizard-of-Oz Dataset for Task-Oriented Dialogue Modelling](https://mlsanity.com/paper/1810.00278/) (2018)
- [Language-agnostic BERT Sentence Embedding](https://mlsanity.com/paper/2007.01852/) (2020)
- [TyDi QA: A Benchmark for Information-Seeking Question Answering in Typologically Diverse Languages](https://mlsanity.com/paper/2003.05002/) (2020)
- [A Simple Language Model for Task-Oriented Dialogue](https://mlsanity.com/paper/2005.00796/) (2020)
- [Neural Approaches to Conversational AI](https://mlsanity.com/paper/1809.08267/) (2018)
- [MTOP: A Comprehensive Multilingual Task-Oriented Semantic Parsing Benchmark](https://mlsanity.com/paper/2008.09335/) (2020)
- [Semantically Conditioned LSTM-based Natural Language Generation for Spoken Dialogue Systems](https://mlsanity.com/paper/1508.01745/) (2015)
- [Cross-Lingual Transfer Learning for Multilingual Task Oriented Dialog](https://mlsanity.com/paper/1810.13327/) (2018)
- [MinTL: Minimalist Transfer Learning for Task-Oriented Dialogue Systems](https://mlsanity.com/paper/2009.12005/) (2020)
- [CrossWOZ: A Large-Scale Chinese Cross-Domain Task-Oriented Dialogue Dataset](https://mlsanity.com/paper/2002.11893/) (2020)
- [BiToD: A Bilingual Multi-Domain Dataset For Task-Oriented Dialogue Modeling](https://mlsanity.com/paper/2106.02787/) (2021)
- [Overview of the Ninth Dialog System Technology Challenge: DSTC9](https://mlsanity.com/paper/2011.06486/) (2020)
- [Semantically Conditioned Dialog Response Generation via Hierarchical Disentangled Self-Attention](https://mlsanity.com/paper/1905.12866/) (2019)
- [RiSAWOZ: A Large-Scale Multi-Domain Wizard-of-Oz Dataset with Rich Semantic Annotations for Task-Oriented Dialogue Modeling](https://mlsanity.com/paper/2010.08738/) (2020)
- [GlobalWoZ: Globalizing MultiWoZ to Develop Multilingual Task-Oriented Dialogue Systems](https://mlsanity.com/paper/2110.07679/) (2021)
- [Multi2WOZ: A Robust Multilingual Dataset and Conversational Pretraining for Task-Oriented Dialog](https://mlsanity.com/paper/2205.10400/) (2022)
- [Large-Scale Multi-Domain Belief Tracking with Knowledge Sharing](https://mlsanity.com/paper/1807.06517/) (2018)
- [Semantic Specialisation of Distributional Word Vector Spaces using Monolingual and Cross-Lingual Constraints](https://mlsanity.com/paper/1706.00374/) (2017)
- [Genie: A Generator of Natural Language Semantic Parsers for Virtual Assistant Commands](https://mlsanity.com/paper/1904.09020/) (2019)
- [Bootstrapping a Crosslingual Semantic Parser](https://mlsanity.com/paper/2004.02585/) (2020)
- [Contextual Semantic Parsing for Multilingual Task-Oriented Dialogues](https://mlsanity.com/paper/2111.02574/) (2021)
- [AllWOZ: Towards Multilingual Task-Oriented Dialog Systems for All](https://mlsanity.com/paper/2112.08333/) (2021)

## Built on by (most cited first, 4 of 4)

- [X-RiSAWOZ: High-Quality End-to-End Multilingual Dialogue Datasets and Few-shot Agents](https://mlsanity.com/paper/2306.17674/) (2023)
- [NextLevelBERT: Masked Language Modeling with Higher-Level Representations for Long Documents](https://mlsanity.com/paper/2402.17682/) (2024)
- [Benchmarks Underestimate the Readiness of Multi-lingual Dialogue Agents](https://mlsanity.com/paper/2405.17840/) (2024)
- [Spec-TOD: A Specialized Instruction-Tuned LLM Framework for Efficient Task-Oriented Dialogue Systems](https://mlsanity.com/paper/2507.04841/) (2025)

## Abstract (arXiv)

Task-oriented Dialogue (ToD) agents are mostly limited to a few widely-spoken languages, mainly due to the high cost of acquiring training data for each language. Existing low-cost approaches that rely on cross-lingual embeddings or naive machine translation sacrifice a lot of accuracy for data efficiency, and largely fail in creating a usable dialogue agent. We propose automatic methods that use ToD training data in a source language to build a high-quality functioning dialogue agent in another target language that has no training data (i.e. zero-shot) or a small training set (i.e. few-shot). Unlike most prior work in cross-lingual ToD that only focuses on Dialogue State Tracking (DST), we build an end-to-end agent. We show that our approach closes the accuracy gap between few-shot and existing full-shot methods for ToD agents. We achieve this by (1) improving the dialogue data representation, (2) improving entity-aware machine translation, and (3) automatic filtering of noisy translations. We evaluate our approach on the recent bilingual dialogue dataset BiToD. In Chinese to English transfer, in the zero-shot setting, our method achieves 46.7% and 22.0% in Task Success Rate (TSR) and Dialogue Success Rate (DSR) respectively. In the few-shot setting where 10% of the data in the target language is used, we improve the state-of-the-art by 15.2% and 14.0%, coming within 5% of full-shot training.