DialUp! Modeling the Language Continuum by Adapting Models to Dialects and Dialects to Models

2025-01-27

AI Summary

Low-resource dialect machine translation suffers from data scarcity and poor cross-variant generalization. Method: We propose a bidirectional adaptation framework: during training, synthetic dialect data augmentation and contrastive learning enable robust “large-language → dialect” (M→D) transfer; during inference, linguistically grounded rewriting and feature disentanglement dynamically calibrate dialect inputs to align with high-resource language models (“dialect → large-language”, D→M). Contribution/Results: This work is the first to systematically model cross-variant generalization across the dialect continuum, balancing robustness to unseen dialects with precise adaptation to known ones. Evaluated across十余 dialects spanning four language families, our approach achieves substantial BLEU gains on low-baseline dialects (+4.2–8.7) and consistent improvements across other families, demonstrating both effectiveness and broad applicability.

Technology Category

• Natural Language Processing: Machine Translation, Multilinguality, Cross-Lingual NLP
• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval

Application Category

• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search
• Web Mining and Content Analysis: Large pretrained models with web data
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation

Abstract

Most of the world's languages and dialects are low-resource, and lack support in mainstream machine translation (MT) models. However, many of them have a closely-related high-resource language (HRL) neighbor, and differ in linguistically regular ways from it. This underscores the importance of model robustness to dialectical variation and cross-lingual generalization to the HRL dialect continuum. We present DialUp, consisting of a training-time technique for adapting a pretrained model to dialectical data (M->D), and an inference-time intervention adapting dialectical data to the model expertise (D->M). M->D induces model robustness to potentially unseen and unknown dialects by exposure to synthetic data exemplifying linguistic mechanisms of dialectical variation, whereas D->M treats dialectical divergence for known target dialects. These methods show considerable performance gains for several dialects from four language families, and modest gains for two other language families. We also conduct feature and error analyses, which show that language varieties with low baseline MT performance are more likely to benefit from these approaches.

Problem

• Machine Translation
• Dialect Adaptation
• Resource-rich to Dialect Transfer

Innovation

• DialUp
• M->D and D->M techniques
• Dialect adaptation in machine translation