Article Per Year (5 Year)
p-Index From 2021 - 2026
2.63
P-Index
This Author published in this journals
All Journal International Journal of Electrical and Computer Engineering IJCCS (Indonesian Journal of Computing and Cybernetics Systems) Seminar Nasional Aplikasi Teknologi Informasi (SNATI) Jurnal Ilmu Komputer dan Informasi Jurnal Informatika Jurnal Teknologi Informasi dan Ilmu Komputer Journal of ICT Research and Applications JOIV : International Journal on Informatics Visualization Jurnal CoreIT CogITo Smart Journal Indonesian Journal of Artificial Intelligence and Data Mining Jurnal Nasional Pendidikan Teknik Informatika (JANAPATI) Jurnal Manajemen Informatika Jurnal Review Pendidikan dan Pengajaran (JRPP) Jurnal Ilmiah Realtech Jurnal Filsafat Indonesia Journal of Applied Data Sciences Malcom: Indonesian Journal of Machine Learning and Computer Science Journal of Multimedia Trend and Technology Jurnal Rekayasa elektrika Jurnal Informatika Polinema (JIP) PROCEEDINGS OF THE INTERNATIONAL CONFERENCE ON DATA SCIENCE AND OFFICIAL STATISTICS
Rinaldi Munir
Bandung Institute of Technology, Jl. Ganesha 10, Bandung 40132, Indonesia
Arts Civil Engineering, Building, Construction & Architecture Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management Economics, Econometrics & Finance Education Electrical & Electronics Engineering Energy Engineering Industrial & Manufacturing Engineering Languange, Linguistic, Communication & Media Library & Information Science Social Sciences Other

Published : 41 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 41 Documents
Search

 1   2   3   4   5 
Comparing Pre-Norm and Post-Norm Transformers in Preserving Gender Information for Indonesian–English Translation through Attention-Based Signal Reinforcement Wijanarko, Andik; Munir, Rinaldi; Khodra, Masayu Leylia; Lestari, Dessi Puji
Journal of Applied Data Sciences Vol 7, No 2: May 2026
Publisher : Bright Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47738/jads.v7i2.1257

Abstract

Gender realization in Indonesian–English machine translation remains challenging due to the absence of grammatical gender in Indonesian, which often leads to unstable or ambiguous gender representations in English outputs. While Transformer-based models have demonstrated strong general translation performance, their ability to preserve gender information across encoding layers remains inconsistent and poorly understood, particularly with respect to architectural normalization strategies.This study presents a comparative analysis of Pre-Norm and Post-Norm Transformer architectures in preserving gender information, and examines the role of attention-based signal reinforcement in mitigating representational degradation. The reinforcement mechanism is introduced prior to standard encoder processing to strengthen gender-relevant token interactions without modifying the overall model structure.Four controlled configurations—Post-Norm, Pre-Norm, Post-Norm with attention-based reinforcement, and Pre-Norm with attention-based reinforcement—are trained under identical random seeds on both unbalanced and balanced datasets. Evaluation is performed on gender-ambiguous test sentences without explicit gender annotations to assess generalization. Gender preservation is assessed at the output level using gender-specific accuracy and BLEU score, and at the representation level using cosine similarity between gender cue embeddings and English gendered pronouns.The results show that Post-Norm Transformers fail to maintain stable gender representations, yielding near-random gender accuracy (~50%) and negligible BLEU scores. Pre-Norm architectures improve training stability but achieve limited gender accuracy (around 30%). Incorporating attention-based signal reinforcement substantially enhances gender preservation, with accuracy rising to over 50% and reaching up to 56% under balanced training conditions, accompanied by a consistent increase in cosine similarity values (exceeding 0.35) between gender cues and corresponding pronouns. These findings indicate that normalization strategy and attention-based reinforcement jointly determine the stability of gender representations in Transformer-based machine translation.
Co-Authors Agus Nursikuwagus Akbar, Saiful Andik Wijanarko, Andik Andreas Prasetia Antonius Sigit Bagus Setyawan Wijaya Bambang Riyanto Daniel Silli Bataona Dessi Puji Lestari, Dessi Puji Dimitri Mahayana Dwi H. Widyantoro Dwi Hendratmo Widyantoro Erry Koryanti Gerung, Altiano R Gloria Christiana Manulangga Hafni Syaeful Sulun Hafni Syaeful Sulun Harlili Harlili Hartanto Tantriawan Hendrana, Hendrana Hendri Hermawan Adinugraha Ignatius Wiseto Prasetyo Agung Imam Habibi Iping Supriana Iping Supriana Suwardi Iping Supriana Suwardi Jemsrado Sine Junaidy Budi Sanger Made Windu Antara Kesiman Masayu Leylia Khodra Menur Wahyu Pangestika, Menur Wahyu Miguel Angelo Lamongi Mohamad Idris Mokhamad Syaroni Nugraha Priya Utama Nur Ulfa Maulidevi Pantouw, Maria L. Permana , John Swatrahadi Pribadi, Fithratulhay Purwoko, Agus Ramdani, Ahmad Luky Ratih Kartika Dewi Rila Mandala Riza Ibnu Adam Sahambangung, Johanis R. Sarwono Sutikno Sarwono Sutikno SATRIYAS ILYAS Sulun, Hafni Syaeful Tedjo Darmanto Tedjo Darmanto Tutun Juhana Utama, Nugraha Priya Wijaya, Bagus Setyawan Zainal Abidin