# Cross-lingual Intermediate Fine-tuning improves Dialogue State Tracking

arXiv: https://arxiv.org/abs/2109.13620 · HTML: https://mlsanity.com/paper/2109.13620/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- Builds on 27 of its 38 references that are in the map.
- First posted on arXiv 28 Sep 2021.

## Builds on (most cited first, 27 of 27)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Don't Stop Pretraining: Adapt Language Models to Domains and Tasks](https://mlsanity.com/paper/2004.10964/) (2020)
- [Cross-lingual Language Model Pretraining](https://mlsanity.com/paper/1901.07291/) (2019)
- [XTREME: A Massively Multilingual Multi-task Benchmark for Evaluating Cross-lingual Generalization](https://mlsanity.com/paper/2003.11080/) (2020)
- [Pre-trained Models for Natural Language Processing: A Survey](https://mlsanity.com/paper/2003.08271/) (2020)
- [Massively Multilingual Sentence Embeddings for Zero-Shot Cross-Lingual Transfer and Beyond](https://mlsanity.com/paper/1812.10464/) (2018)
- [A Network-based End-to-End Trainable Task-oriented Dialogue System](https://mlsanity.com/paper/1604.04562/) (2016)
- [Sentence Encoders on STILTs: Supplementary Training on Intermediate Labeled-data Tasks](https://mlsanity.com/paper/1811.01088/) (2018)
- [Muppet: Massive Multi-task Representations with Pre-Finetuning](https://mlsanity.com/paper/2101.11038/) (2021)
- [TransferTransfo: A Transfer Learning Approach for Neural Network Based Conversational Agents](https://mlsanity.com/paper/1901.08149/) (2019)
- [Neural Belief Tracker: Data-Driven Dialogue State Tracking](https://mlsanity.com/paper/1606.03777/) (2016)
- [Intermediate-Task Transfer Learning with Pretrained Models for Natural Language Understanding: When and Why Does It Work?](https://mlsanity.com/paper/2005.00628/) (2020)
- [Cross-Lingual Transfer Learning for Multilingual Task Oriented Dialog](https://mlsanity.com/paper/1810.13327/) (2018)
- [CoSDA-ML: Multi-Lingual Code-Switching Data Augmentation for Zero-Shot Cross-Lingual NLP](https://mlsanity.com/paper/2006.06402/) (2020)
- [CrossWOZ: A Large-Scale Chinese Cross-Domain Task-Oriented Dialogue Dataset](https://mlsanity.com/paper/2002.11893/) (2020)
- [SUMBT: Slot-Utterance Matching for Universal and Scalable Belief Tracking](https://mlsanity.com/paper/1907.07421/) (2019)
- [ConvLab-2: An Open-Source Toolkit for Building, Evaluating, and Diagnosing Dialogue Systems](https://mlsanity.com/paper/2002.04793/) (2020)
- [Attention-Informed Mixed-Language Training for Zero-shot Cross-lingual Task-oriented Dialogue Systems](https://mlsanity.com/paper/1911.09273/) (2019)
- [On The Evaluation of Machine Translation Systems Trained With Back-Translation](https://mlsanity.com/paper/1908.05204/) (2019)
- [English Intermediate-Task Training Improves Zero-Shot Cross-Lingual Transfer Too](https://mlsanity.com/paper/2005.13013/) (2020)
- [Pretraining Methods for Dialog Context Representation Learning](https://mlsanity.com/paper/1906.00414/) (2019)
- [Is Supervised Syntactic Parsing Beneficial for Language Understanding? An Empirical Investigation](https://mlsanity.com/paper/2008.06788/) (2020)
- [Towards Universal Dialogue State Tracking](https://mlsanity.com/paper/1810.09587/) (2018)
- [Semantic Specialisation of Distributional Word Vector Spaces using Monolingual and Cross-Lingual Constraints](https://mlsanity.com/paper/1706.00374/) (2017)
- [XL-NBT: A Cross-lingual Neural Belief Tracking Framework](https://mlsanity.com/paper/1808.06244/) (2018)
- [An Empirical Study of Cross-Lingual Transferability in Generative Dialogue State Tracker](https://mlsanity.com/paper/2101.11360/) (2021)

## Built on by (most cited first, 11 of 11)

- [LIFT: Language-Interfaced Fine-Tuning for Non-Language Machine Learning Tasks](https://mlsanity.com/paper/2206.06565/) (2022)
- [Improving In-Context Few-Shot Learning via Self-Supervised Training](https://mlsanity.com/paper/2205.01703/) (2022)
- [Multi2WOZ: A Robust Multilingual Dataset and Conversational Pretraining for Task-Oriented Dialog](https://mlsanity.com/paper/2205.10400/) (2022)
- ["Do you follow me?": A Survey of Recent Approaches in Dialogue State Tracking](https://mlsanity.com/paper/2207.14627/) (2022)
- [MULTI3NLU++: A Multilingual, Multi-Intent, Multi-Domain Dataset for Natural Language Understanding in Task-Oriented Dialogue](https://mlsanity.com/paper/2212.10455/) (2022)
- [X-RiSAWOZ: High-Quality End-to-End Multilingual Dialogue Datasets and Few-shot Agents](https://mlsanity.com/paper/2306.17674/) (2023)
- [XQA-DST: Multi-Domain and Multi-Lingual Dialogue State Tracking](https://mlsanity.com/paper/2204.05895/) (2022)
- [Exploiting Domain-Specific Parallel Data on Multilingual Language Models for Low-resource Language Translation](https://mlsanity.com/paper/2412.19522/) (2024)
- [A Major Obstacle for NLP Research: Let's Talk about Time Allocation!](https://mlsanity.com/paper/2211.16858/) (2022)
- [Enhancing the Preference Extractor in Multi-turn Dialogues: From Annotating Disasters to Accurate Preference Extraction](https://mlsanity.com/paper/2508.01739/) (2025)
- [Building Dialogue Understanding Models for Low-resource Language Indonesian from Scratch](https://mlsanity.com/paper/2410.18430/) (2024)

## Abstract (arXiv)

Recent progress in task-oriented neural dialogue systems is largely focused on a handful of languages, as annotation of training data is tedious and expensive. Machine translation has been used to make systems multilingual, but this can introduce a pipeline of errors. Another promising solution is using cross-lingual transfer learning through pretrained multilingual models. Existing methods train multilingual models with additional code-mixed task data or refine the cross-lingual representations through parallel ontologies. In this work, we enhance the transfer learning process by intermediate fine-tuning of pretrained multilingual models, where the multilingual models are fine-tuned with different but related data and/or tasks. Specifically, we use parallel and conversational movie subtitles datasets to design cross-lingual intermediate tasks suitable for downstream dialogue tasks. We use only 200K lines of parallel data for intermediate fine-tuning which is already available for 1782 language pairs. We test our approach on the cross-lingual dialogue state tracking task for the parallel MultiWoZ (English -> Chinese, Chinese -> English) and Multilingual WoZ (English -> German, English -> Italian) datasets. We achieve impressive improvements (> 20% on joint goal accuracy) on the parallel MultiWoZ dataset and the Multilingual WoZ dataset over the vanilla baseline with only 10% of the target language task data and zero-shot setup respectively.