paper-with-me

Papers

Understanding the effects of language-specific class imbalance in multilingual fine-tuning

2024-02-20 · Vincent Jung, Lonneke van der Plas

We study the effect of one type of imbalance often present in real-life multilingual classification datasets: an uneven distribution of labels across languages. We show evidence that fine-tuning a transformer-based Large Language Model (LLM) on a dataset with this imbalance leads to worse performance, a more pronounced separation of languages in the latent space, and the promotion of uninformative features. We modify the traditional class weighing approach to imbalance by calculating class weights separately for each language and show that this helps mitigate those detrimental effects. These results create awareness of the negative effects of language-specific class imbalance in multilingual fine-tuning and the way in which the model learns to rely on the separation of languages to perform the task.

📄 PDF Abstract BibTeX arXiv:2402.13016

Code (1)

idiap/class-imbalance-multilingual-ft 공식 구현 pytorch

Tasks

Language ModelingLanguage ModellingLarge Language Model

Similar Papers 제목 키워드 기반

Imbalanced Classification through the Lens of Spurious Correlations

2025-10-31 · Jakob Hackstein, Sidney Bender arxiv

Class imbalance poses a fundamental challenge in machine learning, frequently leading to unreliable classification performance. While prior methods focus on data- or loss-reweighting schemes, we view imbalance as a data …

Balancing the Causal Effects in Class-Incremental Learning

2024-02-15 · Junhao Zheng, Ruiyan Wang, Chongzhi Zhang, Huawen Feng 외

Class-Incremental Learning (CIL) is a practical and challenging problem for achieving general artificial intelligence. Recently, Pre-Trained Models (PTMs) have led to breakthroughs in both visual and natural language pro…

class-incremental learningClass Incremental LearningContinual Named Entity Recognitionimage-classification+6

Biased Minds Meet Biased AI: How Class Imbalance Shapes Appropriate Reliance and Interacts with Human Base Rate Neglect

2025-11-18 · Nick von Felten, Johannes Schöning, Klaus Opwis, Nicolas Scharowski arxiv

Humans increasingly interact with artificial intelligence (AI) in decision-making. However, both AI and humans are prone to biases. While AI and human biases have been studied extensively in isolation, this paper examine…

Class-Dependent Perturbation Effects in Evaluating Time Series Attributions

2025-02-24 · Gregor Baer, Isel Grau, Chao Zhang, Pieter Van Gorp

As machine learning models become increasingly prevalent in time series applications, Explainable Artificial Intelligence (XAI) methods are essential for understanding their predictions. Within XAI, feature attribution m…

Explainable artificial intelligenceExplainable Artificial Intelligence (XAI)Time SeriesTime Series Classification

Sentence Classification with Imbalanced Data for Health Applications

2020-12-01 · SMM4H (COLING) 2020 12 · Farhana Ferdousi Liza

Identifying and extracting reports of medications, their abuse or adverse effects from social media is a challenging task. In social media, relevant reports are very infrequent, causes imbalanced class distribution for m…

ClassificationSentenceSentence ClassificationTask 2