@inproceedings{dai-etal-2026-metacora,
title = "{M}eta{CORA}: A Meta-Learned Curriculum for Adversarial and Contrastive Robustness in Speech Recognition",
author = "Dai, Yuqian and
Chan, Chun Fai and
Wong, Ying Ki and
Pun, Tsz Ho",
editor = "Piperidis, Stelios and
Bel, N{\'u}ria and
van den Heuvel, Henk and
Ide, Nancy and
Krek, Simon and
Toral, Antonio",
booktitle = "Proceedings of the Fifteenth Language Resources and Evaluation Conference",
month = may,
year = "2026",
address = "Palma de Mallorca, Spain",
publisher = "ELRA Language Resource Association",
url = "https://aclanthology.org/2026.lrec-1.784/",
doi = "10.63317/55npni82hjf5",
pages = "9999--10008",
abstract = "Pre-trained speech models like Whisper demonstrate impressive performance under ideal conditions but still face robustness challenges in low-resource language scenarios. We introduce Meta Curriculum Optimization for Robust ASR (MetaCORA), a novel meta-curriculum adaptive framework that improves speech recognition for low-resource Hong Kong Cantonese by integrating adversarial training with feature contrastive learning. Our approach dynamically adjusts three critical hyperparameters: adversarial perturbation magnitude, optimization step size, and contrastive learning temperature, allowing the model to adapt to varying training difficulties throughout the learning process. Unlike traditional meta-learning approaches, our framework does not rely on end-to-end differentiability but instead utilizes validation performance as a signal to guide hyperparameter adjustments. Experimental results demonstrate that our approach achieves lower WER than standard Whisper fine-tuning, commercial speech recognition systems, and LLM-based methods. Ablation studies confirm the necessity of each component, as removing any single element leads to a measurable drop in performance. The model also exhibits robustness under noisy conditions, achieving consistently lower WER than baseline systems. Further analysis shows that MetaCORA effectively compresses the distance between adversarial feature representations while maintaining well-separated class boundaries in the embedding space, providing a mechanistic explanation for its improvement."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="dai-etal-2026-metacora">
<titleInfo>
<title>MetaCORA: A Meta-Learned Curriculum for Adversarial and Contrastive Robustness in Speech Recognition</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yuqian</namePart>
<namePart type="family">Dai</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chun</namePart>
<namePart type="given">Fai</namePart>
<namePart type="family">Chan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ying</namePart>
<namePart type="given">Ki</namePart>
<namePart type="family">Wong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tsz</namePart>
<namePart type="given">Ho</namePart>
<namePart type="family">Pun</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-05</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the Fifteenth Language Resources and Evaluation Conference</title>
</titleInfo>
<name type="personal">
<namePart type="given">Stelios</namePart>
<namePart type="family">Piperidis</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Núria</namePart>
<namePart type="family">Bel</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Henk</namePart>
<namePart type="family">van den Heuvel</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nancy</namePart>
<namePart type="family">Ide</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Simon</namePart>
<namePart type="family">Krek</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Antonio</namePart>
<namePart type="family">Toral</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>ELRA Language Resource Association</publisher>
<place>
<placeTerm type="text">Palma de Mallorca, Spain</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Pre-trained speech models like Whisper demonstrate impressive performance under ideal conditions but still face robustness challenges in low-resource language scenarios. We introduce Meta Curriculum Optimization for Robust ASR (MetaCORA), a novel meta-curriculum adaptive framework that improves speech recognition for low-resource Hong Kong Cantonese by integrating adversarial training with feature contrastive learning. Our approach dynamically adjusts three critical hyperparameters: adversarial perturbation magnitude, optimization step size, and contrastive learning temperature, allowing the model to adapt to varying training difficulties throughout the learning process. Unlike traditional meta-learning approaches, our framework does not rely on end-to-end differentiability but instead utilizes validation performance as a signal to guide hyperparameter adjustments. Experimental results demonstrate that our approach achieves lower WER than standard Whisper fine-tuning, commercial speech recognition systems, and LLM-based methods. Ablation studies confirm the necessity of each component, as removing any single element leads to a measurable drop in performance. The model also exhibits robustness under noisy conditions, achieving consistently lower WER than baseline systems. Further analysis shows that MetaCORA effectively compresses the distance between adversarial feature representations while maintaining well-separated class boundaries in the embedding space, providing a mechanistic explanation for its improvement.</abstract>
<identifier type="citekey">dai-etal-2026-metacora</identifier>
<identifier type="doi">10.63317/55npni82hjf5</identifier>
<location>
<url>https://aclanthology.org/2026.lrec-1.784/</url>
</location>
<part>
<date>2026-05</date>
<extent unit="page">
<start>9999</start>
<end>10008</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T MetaCORA: A Meta-Learned Curriculum for Adversarial and Contrastive Robustness in Speech Recognition
%A Dai, Yuqian
%A Chan, Chun Fai
%A Wong, Ying Ki
%A Pun, Tsz Ho
%Y Piperidis, Stelios
%Y Bel, Núria
%Y van den Heuvel, Henk
%Y Ide, Nancy
%Y Krek, Simon
%Y Toral, Antonio
%S Proceedings of the Fifteenth Language Resources and Evaluation Conference
%D 2026
%8 May
%I ELRA Language Resource Association
%C Palma de Mallorca, Spain
%F dai-etal-2026-metacora
%X Pre-trained speech models like Whisper demonstrate impressive performance under ideal conditions but still face robustness challenges in low-resource language scenarios. We introduce Meta Curriculum Optimization for Robust ASR (MetaCORA), a novel meta-curriculum adaptive framework that improves speech recognition for low-resource Hong Kong Cantonese by integrating adversarial training with feature contrastive learning. Our approach dynamically adjusts three critical hyperparameters: adversarial perturbation magnitude, optimization step size, and contrastive learning temperature, allowing the model to adapt to varying training difficulties throughout the learning process. Unlike traditional meta-learning approaches, our framework does not rely on end-to-end differentiability but instead utilizes validation performance as a signal to guide hyperparameter adjustments. Experimental results demonstrate that our approach achieves lower WER than standard Whisper fine-tuning, commercial speech recognition systems, and LLM-based methods. Ablation studies confirm the necessity of each component, as removing any single element leads to a measurable drop in performance. The model also exhibits robustness under noisy conditions, achieving consistently lower WER than baseline systems. Further analysis shows that MetaCORA effectively compresses the distance between adversarial feature representations while maintaining well-separated class boundaries in the embedding space, providing a mechanistic explanation for its improvement.
%R 10.63317/55npni82hjf5
%U https://aclanthology.org/2026.lrec-1.784/
%U https://doi.org/10.63317/55npni82hjf5
%P 9999-10008
Markdown (Informal)
[MetaCORA: A Meta-Learned Curriculum for Adversarial and Contrastive Robustness in Speech Recognition](https://aclanthology.org/2026.lrec-1.784/) (Dai et al., LREC 2026)
ACL