MultiMatch: Multihead Consistency Regularization Matching for Semi-Supervised Text Classification

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Sirbu, Iustin, Popovici, Robert-Adrian, Caragea, Cornelia, Trausan-Matu, Stefan, Rebedea, Traian
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914129077338112
author Sirbu, Iustin
Popovici, Robert-Adrian
Caragea, Cornelia
Trausan-Matu, Stefan
Rebedea, Traian
author_facet Sirbu, Iustin
Popovici, Robert-Adrian
Caragea, Cornelia
Trausan-Matu, Stefan
Rebedea, Traian
contents We introduce MultiMatch, a novel semi-supervised learning (SSL) algorithm combining the paradigms of co-training and consistency regularization with pseudo-labeling. At its core, MultiMatch features a pseudo-label weighting module designed for selecting and filtering pseudo-labels based on head agreement and model confidence, and weighting them according to the perceived classification difficulty. This novel module enhances and unifies three existing techniques -- heads agreement from Multihead Co-training, self-adaptive thresholds from FreeMatch, and Average Pseudo-Margins from MarginMatch -- resulting in a holistic approach that improves robustness and performance in SSL settings. Experimental results on benchmark datasets highlight the superior performance of MultiMatch, i.e., MultiMatch achieves state-of-the-art results on 8 out of 10 setups from 5 natural language processing datasets and ranks first according to the Friedman test among 21 methods. Furthermore, MultiMatch demonstrates exceptional robustness in highly imbalanced settings, outperforming the second-best approach by 3.26%, a critical advantage for real-world text classification tasks. Our code is available on GitHub.
format Preprint
id arxiv_https___arxiv_org_abs_2506_07801
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle MultiMatch: Multihead Consistency Regularization Matching for Semi-Supervised Text Classification
Sirbu, Iustin
Popovici, Robert-Adrian
Caragea, Cornelia
Trausan-Matu, Stefan
Rebedea, Traian
Computation and Language
Artificial Intelligence
Machine Learning
I.2.7
We introduce MultiMatch, a novel semi-supervised learning (SSL) algorithm combining the paradigms of co-training and consistency regularization with pseudo-labeling. At its core, MultiMatch features a pseudo-label weighting module designed for selecting and filtering pseudo-labels based on head agreement and model confidence, and weighting them according to the perceived classification difficulty. This novel module enhances and unifies three existing techniques -- heads agreement from Multihead Co-training, self-adaptive thresholds from FreeMatch, and Average Pseudo-Margins from MarginMatch -- resulting in a holistic approach that improves robustness and performance in SSL settings. Experimental results on benchmark datasets highlight the superior performance of MultiMatch, i.e., MultiMatch achieves state-of-the-art results on 8 out of 10 setups from 5 natural language processing datasets and ranks first according to the Friedman test among 21 methods. Furthermore, MultiMatch demonstrates exceptional robustness in highly imbalanced settings, outperforming the second-best approach by 3.26%, a critical advantage for real-world text classification tasks. Our code is available on GitHub.
title MultiMatch: Multihead Consistency Regularization Matching for Semi-Supervised Text Classification
topic Computation and Language
Artificial Intelligence
Machine Learning
I.2.7
url https://arxiv.org/abs/2506.07801