SentiMatrix: Parameter-Efficient Fine-Tuning of Encoder-Based Transformers for Multidimensional Sentiment Analysis
Machine Learning · 2026-10 · 115 (10)
DOI: 10.1007/s10994-026-07161-4 ↗Kurz gefasst
Fachartikel in der Zeitschrift Machine Learning, online erschienen am 24.09.2026. Verfasst von Md. Easin Arafat und 3 weiteren Autor:innen. Die Arbeit nennt 63 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Eötvös Loránd University.
Abstract (Originalsprache)
Transformer-based architectures have established state-of-the-art benchmarks across Natural Language Processing (NLP) tasks; however, the computational overhead of full fine-tuning remains a significant barrier to scalable deployment. This paper presents SentiMatrix, a systematic evaluation of Low-Rank Adaptation (LoRA)-based Parameter-Efficient Fine-Tuning (PEFT) across four sentiment analysis paradigms: Intent-based, Aspect-based, Fine-grained, and Emotion detection, spanning seven benchmark datasets under a consistent three-stage protocol: inference using task-specific pretrained baseline models (i.e., HuggingFace checkpoints previously fine-tuned on the corresponding benchmarks, serving as informed upper-bound references rather than cold-start zero-shot baselines), Full Fine-Tuning (FFT) as a performance upper bound, and LoRA-based adaptation to quantify the efficiency-performance trade-off. Adaptive Low-Rank Adaptation (AdaLoRA) is benchmarked as a direct comparator. LoRA achieves competitive performance while reducing trainable parameters by up to 99.8%, with lower GPU memory usage and training time in most evaluated settings. It achieves 93.28% accuracy on SST-2, 80.74% on combined Laptop+Restaurant ABSA, 67.63% on E-Commerce 5-class, and 93.22% on emotion detection, while closely approaching FFT on E-Commerce 5-class (68.47%). The results demonstrate an effective efficiency-performance trade-off across diverse sentiment analysis tasks. Ablation experiments examine the effects of adaptation strategy, model capacity, and label granularity. The source code and datasets are available at https://github.com/ELTE-DSED/senti-matrix .
Förderung
- Eötvös Loránd University
Zitieren
APA
Arafat, M. E., Akmal, M. U., Abosinnee, A. S., & Orosz, T. (2026). SentiMatrix: Parameter-Efficient Fine-Tuning of Encoder-Based Transformers for Multidimensional Sentiment Analysis. Machine Learning, 115(10). https://doi.org/10.1007/s10994-026-07161-4
Vancouver
Arafat ME, Akmal MU, Abosinnee AS, Orosz T. SentiMatrix: Parameter-Efficient Fine-Tuning of Encoder-Based Transformers for Multidimensional Sentiment Analysis. Machine Learning. 2026;115(10). doi:10.1007/s10994-026-07161-4
BibTeX
@article{Arafat2026,
author = {Arafat, Md. Easin and Akmal, Muhammad Usman and Abosinnee, Ali S. and Orosz, Tamás},
title = {SentiMatrix: Parameter-Efficient Fine-Tuning of Encoder-Based Transformers for Multidimensional Sentiment Analysis},
journal = {Machine Learning},
year = {2026},
volume = {115},
number = {10},
doi = {10.1007/s10994-026-07161-4}
}
EUVIMED