・Peer-Reviewed International Conference Proceedings
1. A. Kojima_, Y. Fujita, H. Shi, T. Mizumoto, M. Zhao, and Y. Sudo, "Conversation Context-aware Direct Preference Optimization for Style-Controlled Speech Synthesis," in Proc. of Asia Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA 2025), 2025.
2. A. Kojima, “Sub-table rescorer for table question answering,” in Proc. of
LREC-COLING 2024, 2024.
3. S. Hiai, S. Nagayama and A. Kojima, “AMI team at the NTCIR-16 Real-MedNLP
Task,” in Proc. of 16th NTCIR Conference on Evaluation of Information Access
Technologie, 2022. (won 1st place in 2 few-resource named entity recognition tasks)
4. A. Kojima, “Large-context automatic speech recognition based on RNN-transducer,”
in Proc. of Asia Pacific Signal and Information Processing Association Annual
Summit and Conference (APSIPA 2021), 2021.
5. A. Kojima, “Knowledge distillation for streaming Transformer–Transducer,” in Proc. of
INTERSPEECH 2021, 2021.
6. A.Kojima, K. Itou, and H. Hanaizumi, “Detection of focus for importance of words:
toward computer-assisted learning for improving presentation,” Symposium on
Information and Communication Technology, in Proc. of International Conference on
Language, Innovation, Culture, and Education (ICLICE 2016), 2016.
7. A.Kojima, K. Itou, “Prominence detection for presentation training system,”
Symposium on Information and Communication Technology, in Proc. of Symposium
on Information and Communication Technology (SoICT 2016), 2016. (recommended
paper of SoICT 2016)
・Domestic Workshop Proceedings
1. A. Kojima, “Intermediate direct preference optimization,” Special interest group of
natural language processing, 2024.
2. A. Kojima, I. Kitagishi, “Houou: Large language model trained by riken
ichikara-instruction dataset,” Annual meeting of the association for natural language
processing, 2024.
3. A. Kojima, “Automated generation of reply emails in customer support based on
retrieval-augmented generation,” Annual meeting of the association for natural
language processing, 2024.
4. S. Sekine, A. Kojima, K. Sadamitsu, I. Kitagishi, “Comparison and analysis of gpt-4
automatic evaluation and human evaluation,”Annual meeting of the association for
natural language processing, 2024.
5. A. Kojima, “Sub-table rescorer for table question answering,” Special interest group
of natural language processing, 2023.
6. A. Kojima, “Self-supervised learning for multi-channel neural transducer,” ASJ
National Convention, 2022.
7. A. Kojima, “Large-context automatic speech recognition using RNN-T and
Transformer,” ASJ National Convention, 2021.
8. A. Kojima, “Study on biasing medical term using WFST for medical speech
recognition,” ASJ National Convention, 2021.
9. A. Kojima, R. Oguro, “Simultaneous realization automatic speech recognition and
speaker diarization using Sequence-to-sequence on Japanese medical
conversation,” ASJ National Convention, 2020.
z10. A. Kojima, R. Oguro, “Multi task learning for designing low latency mask-based
MVDR beamformer,” ASJ National Convention, 2020.
11. A. Kojima, K. Itou, “Prominence detection in a presentation using an accent
component,” IPSJ National Convention, 2017. (Student Encouragement Award of
IPSJ National Convention)
12. A. Kojima, K. Itou, “Emotion expression analysis in acted speech using ∆accent,”
ASJ National Convention, 2017.
13. R. Eda, A.Kojima and K.itou, “Speech-rate conversion system to hold the emotional
impression of the speaker,” IPSJ National Convention, 2017.
14. D. Tanaka, A. Kojima and K. Itou, “An automatically arrangement system for
electronic organ scores from multi-part musical scores by clustering of role of
melody,” IPSJ National Convention, 2017. (Student Encouragement Award of IPSJ
National Convention)
15. K. Sato, A. Kojima and K. Itou, “Fill-in conversation using grouping structure and
metrical structure for the drum performance beginners,” IPSJ National Convention,
2017.
16. A. Kojima, K. Itou and H. Hanaizumi, “Detecting an important emphasis in
presentation speech using maximum F0 and maximum accent components,” IPSJ
National Convention, 2016.
17. A. Kojima, K. Itou, “Computer aided training for acquisition of focus on presentation,”
IPSJ National Convention, 2015.
・Article
1. A. Kojima, “A study of biasing technical terms in medical speech recognition using
weighted finite state transducer,” Acoustical Science and Technology, Vol. 43, no. 1,
2021. (Invited article)
・Research talk
1. “Minimum word error rate training for self-conditioned CTC,” 50th Anniversary event
of ASJ-tokai, 2023.
2. “Lightning talk for acoustics (automatic speech recognition),” ASJ National
Convention beginners seminar, 2022.
・Technical blog
1. “Advancing financial AI: Money Forward Lab's approach to specialized LLMs in
Japan,”, https://moneyforward-dev.jp/entry/2024/08/29/124453, 2024. (Japanese)
2. “Houou: large language model trained by Japanese instruction data created from
scratch, ”https://global.moneyforward-dev.jp/2024/01/25/large-language-model-trained
-by-houou/, 2023. (Translated by tech PR div.)