# Mitigating Language-Level Performance Disparity in mPLMs via Teacher Language Selection and Cross-lingual Self-Distillation

arXiv: https://arxiv.org/abs/2404.08491 · HTML: https://mlsanity.com/paper/2404.08491/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Knowledge distillation for LLMs, one of 12 directions in Knowledge distillation.
- Builds on 32 of its 47 references that are in the map.
- First posted on arXiv 12 Apr 2024.

## Placed in

- Knowledge distillation / [Knowledge distillation for LLMs](https://mlsanity.com/distill/knowledge-distillation-for-llms/)

## Builds on (most cited first, 30 of 32)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Unsupervised Cross-lingual Representation Learning at Scale](https://mlsanity.com/paper/1911.02116/) (2019)
- [mT5: A massively multilingual pre-trained text-to-text transformer](https://mlsanity.com/paper/2010.11934/) (2020)
- [XNLI: Evaluating Cross-lingual Sentence Representations](https://mlsanity.com/paper/1809.05053/) (2018)
- [It's Not Just Size That Matters: Small Language Models Are Also Few-Shot Learners](https://mlsanity.com/paper/2009.07118/) (2020)
- [MLQA: Evaluating Cross-lingual Extractive Question Answering](https://mlsanity.com/paper/1910.07475/) (2019)
- [XCOPA: A Multilingual Dataset for Causal Commonsense Reasoning](https://mlsanity.com/paper/2005.00333/) (2020)
- [Improving Massively Multilingual Neural Machine Translation and Zero-Shot Translation](https://mlsanity.com/paper/2004.11867/) (2020)
- [PAWS-X: A Cross-lingual Adversarial Dataset for Paraphrase Identification](https://mlsanity.com/paper/1908.11828/) (2019)
- [Negated and Misprimed Probes for Pretrained Language Models: Birds Can Talk, But Cannot Fly](https://mlsanity.com/paper/1911.03343/) (2019)
- [InfoXLM: An Information-Theoretic Framework for Cross-Lingual Language Model Pre-Training](https://mlsanity.com/paper/2007.07834/) (2020)
- [Generating Training Data with Language Models: Towards Zero-Shot Language Understanding](https://mlsanity.com/paper/2202.04538/) (2022)
- [On the Effectiveness of Adapter-based Tuning for Pretrained Language Model Adaptation](https://mlsanity.com/paper/2106.03164/) (2021)
- [Multilingual LAMA: Investigating Knowledge in Multilingual Pretrained Language Models](https://mlsanity.com/paper/2102.00894/) (2021)
- [XLM-E: Cross-lingual Language Model Pre-training via ELECTRA](https://mlsanity.com/paper/2106.16138/) (2021)
- [SpokenWOZ: A Large-Scale Speech-Text Benchmark for Spoken Task-Oriented Dialogue Agents](https://mlsanity.com/paper/2305.13040/) (2023)
- [ML-Bench: Evaluating Large Language Models and Agents for Machine Learning Tasks on Repository-Level Code](https://mlsanity.com/paper/2311.09835/) (2023)
- [ERNIE-M: Enhanced Multilingual Representation by Aligning Cross-lingual Semantics with Monolingual Corpora](https://mlsanity.com/paper/2012.15674/) (2020)
- [Explicit Alignment Objectives for Multilingual Bidirectional Encoders](https://mlsanity.com/paper/2010.07972/) (2020)
- [A Primer on Pretrained Multilingual Language Models](https://mlsanity.com/paper/2107.00676/) (2021)
- [Improving Pretrained Cross-Lingual Language Models via Self-Labeled Word Alignment](https://mlsanity.com/paper/2106.06381/) (2021)
- [BERTnesia: Investigating the capture and forgetting of knowledge in BERT](https://mlsanity.com/paper/2106.02902/) (2021)
- [VECO: Variable and Flexible Cross-lingual Pre-training for Language Understanding and Generation](https://mlsanity.com/paper/2010.16046/) (2020)
- [Recipes for Adapting Pre-trained Monolingual and Multilingual Models to Machine Translation](https://mlsanity.com/paper/2004.14911/) (2020)
- [SCL-RAI: Span-based Contrastive Learning with Retrieval Augmented Inference for Unlabeled Entity Problem in NER](https://mlsanity.com/paper/2209.01646/) (2022)
- [Mining Clues from Incomplete Utterance: A Query-enhanced Network for Incomplete Utterance Rewriting](https://mlsanity.com/paper/2307.00866/) (2023)
- [Coarse-to-Fine Dual Encoders are Better Frame Identification Learners](https://mlsanity.com/paper/2310.13316/) (2023)
- [Cross-Lingual Text Classification with Multilingual Distillation and Zero-Shot-Aware Training](https://mlsanity.com/paper/2202.13654/) (2022)
- [UniPCM: Universal Pre-trained Conversation Model with Task-aware Automatic Prompt](https://mlsanity.com/paper/2309.11065/) (2023)

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

- [Looking Beyond Text: Reducing Language bias in Large Vision-Language Models via Multimodal Dual-Attention and Soft-Image Guidance](https://mlsanity.com/paper/2411.14279/) (2024)

## Abstract (arXiv)

Large-scale multilingual Pretrained Language Models (mPLMs) yield impressive performance on cross-language tasks, yet significant performance disparities exist across different languages within the same mPLM. Previous studies endeavored to narrow these disparities by supervise fine-tuning the mPLMs with multilingual data. However, obtaining labeled multilingual data is time-consuming, and fine-tuning mPLM with limited labeled multilingual data merely encapsulates the knowledge specific to the labeled data. Therefore, we introduce ALSACE to leverage the learned knowledge from the well-performing languages to guide under-performing ones within the same mPLM, eliminating the need for additional labeled multilingual data. Experiments show that ALSACE effectively mitigates language-level performance disparity across various mPLMs while showing the competitive performance on different multilingual NLU tasks, ranging from full resource to limited resource settings. The code for our approach is available at https://github.com/pkunlp-icler/ALSACE.