Masayuki Goto
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Masayuki Goto has a total of 83 co-authors. Here are the top co-authors by number of shared publications:
- Ryotaro Shimizu
Institute of Data Science, Waseda University, Tokyo 169-8555, Japan
10 shared publications - Linxin Song
Waseda University
3 shared publications - Tengfei Shao
Waseda University
3 shared publications - Tokimasa Isomura
Graduate School of Creative Science and Engineering, Waseda University, Tokyo, Japan
3 shared publications - Ao Luo
KDDI Research, Inc
2 shared publications Haruka Yamashita
Department of Information and Communication Sciences Sophia University Tokyo Japan
2 shared publications
Latest Publications
- Open Access
- 5 November 2026
- Published by Taylor & Francis in Journal of Industrial and Production Engineering
- p. 1-23
Evaluating Wasserstein-robust inventory decisions: a five-gate framework and a plant-level coal inventory study
- 5 November 2026
- Published by Taylor & Francis in Journal of Industrial and Production Engineering
- p. 1-23
Open AccessEvaluating Wasserstein-robust inventory decisions: a five-gate framework and a plant-level coal inventory study
- Open Access
- 24 June 2026
- Published by MDPI AG in Administrative Sciences
- Vol. 16 (7) , 307
Impact of Behavioral Change on Performance Improvement: A Longitudinal Analysis Using Personnel Appraisal Data
This study examines whether within-person changes in behavioral evaluation factors are associated with performance improvement using longitudinal personnel appraisal data from a Japanese firm (fiscal years 2019–2021, N = 218). Fourteen behavioral appraisal items were standardized annually and analyzed using exploratory factor analysis, yielding three factors: Comprehensive Behavior, Job Skills, and Discipline/Values. Behavioral change was defined as the difference in factor scores between FY2021 and FY2019. Hierarchical cluster analysis identified four typical behavioral change patterns. Binary logistic regression results indicate that changes in Job Skills are significantly associated with a higher likelihood of performance improvement (odds ratio = 1.33, 95% confidence interval = 1.02–1.75, p = 0.037). Additional sensitivity analyses controlling for baseline performance and baseline factor levels supported the positive association between changes in Job Skills and performance improvement. These findings suggest that improvement in job-related skills, rather than overall behavioral level, is associated with short- to medium-term performance improvement. The study provides a change-based analytical framework that leverages longitudinal appraisal data to identify behavioral domains linked to performance improvement, thereby contributing to data-driven human resource management and employee development.
- 24 June 2026
- Published by MDPI AG in Administrative Sciences
- Vol. 16 (7) , 307
Open AccessImpact of Behavioral Change on Performance Improvement: A Longitudinal Analysis Using Personnel Appraisal Data
This study examines whether within-person changes in behavioral evaluation factors are associated with performance improvement using longitudinal personnel appraisal data from a Japanese firm (fiscal years 2019–2021, N = 218). Fourteen behavioral appraisal items were standardized annually and analyzed using exploratory factor analysis, yielding three factors: Comprehensive Behavior, Job Skills, and Discipline/Values. Behavioral change was defined as the difference in factor scores between FY2021 and FY2019. Hierarchical cluster analysis identified four typical behavioral change patterns. Binary logistic regression results indicate that changes in Job Skills are significantly associated with a higher likelihood of performance improvement (odds ratio = 1.33, 95% confidence interval = 1.02–1.75, p = 0.037). Additional sensitivity analyses controlling for baseline performance and baseline factor levels supported the positive association between changes in Job Skills and performance improvement. These findings suggest that improvement in job-related skills, rather than overall behavioral level, is associated with short- to medium-term performance improvement. The study provides a change-based analytical framework that leverages longitudinal appraisal data to identify behavioral domains linked to performance improvement, thereby contributing to data-driven human resource management and employee development.
- Open Access
- 20 June 2026
- Published by Atlantis Press SARL
- p. 13-28
A Field Survey on Conditions of Public Perceptions toward Basic Infrastructure in Nepal
- 20 June 2026
- Published by Atlantis Press SARL
- p. 13-28
Open AccessA Field Survey on Conditions of Public Perceptions toward Basic Infrastructure in Nepal
- Open Access
- 14 May 2026
- Published by Atlantis Press SARL
- p. 148-165
Analysis of Effectiveness of Active Learning based Collaborative Educational Program among Universities: A Case Study of Nepal-Japan Project
- 14 May 2026
- Published by Atlantis Press SARL
- p. 148-165
Open AccessAnalysis of Effectiveness of Active Learning based Collaborative Educational Program among Universities: A Case Study of Nepal-Japan Project
- Top Cited PapersOpen Access
- 24 March 2026
- Published by MDPI AG in Journal of Imaging
- Vol. 12 (4) , 143
Vision and Language Reference for a Segment Anything Model for Few-Shot Segmentation
Segment Anything Model (SAM)-based few-shot segmentation models traditionally rely solely on annotated reference images as prompts, which inherently limits their accuracy due to an over-reliance on visual cues and a lack of semantic context. This reliance leads to incorrect segmentation, where visually similar objects from different categories are incorrectly identified as the target object. We propose Vision and Language Reference Prompt into SAM (VLP-SAM), a novel few-shot segmentation model that integrates both visual information of reference images and semantic information of text labels into SAM. VLP-SAM introduces a vision-language model (VLM) with pixel–text matching into the prompt encoder for SAM, effectively leveraging textual semantic consistency while preserving SAM’s extensive segmentation knowledge. By incorporating task-specific structures such as an attention mask, our model achieves superior few-shot segmentation performance with only 1.4 M learnable parameters. Evaluations on PASCAL-5i and COCO-20i datasets demonstrate that VLP-SAM significantly outperforms previous methods by 6.8% and 9.3% in mIoU, respectively. Furthermore, VLP-SAM exhibits strong generalization across unseen objects and cross-domain scenarios, highlighting the robustness provided by textual semantic guidance. This study offers an effective and scalable framework for few-shot segmentation with multimodal prompts.
- 24 March 2026
- Published by MDPI AG in Journal of Imaging
- Vol. 12 (4) , 143
Top Cited PapersOpen AccessVision and Language Reference for a Segment Anything Model for Few-Shot Segmentation
Segment Anything Model (SAM)-based few-shot segmentation models traditionally rely solely on annotated reference images as prompts, which inherently limits their accuracy due to an over-reliance on visual cues and a lack of semantic context. This reliance leads to incorrect segmentation, where visually similar objects from different categories are incorrectly identified as the target object. We propose Vision and Language Reference Prompt into SAM (VLP-SAM), a novel few-shot segmentation model that integrates both visual information of reference images and semantic information of text labels into SAM. VLP-SAM introduces a vision-language model (VLM) with pixel–text matching into the prompt encoder for SAM, effectively leveraging textual semantic consistency while preserving SAM’s extensive segmentation knowledge. By incorporating task-specific structures such as an attention mask, our model achieves superior few-shot segmentation performance with only 1.4 M learnable parameters. Evaluations on PASCAL-5i and COCO-20i datasets demonstrate that VLP-SAM significantly outperforms previous methods by 6.8% and 9.3% in mIoU, respectively. Furthermore, VLP-SAM exhibits strong generalization across unseen objects and cross-domain scenarios, highlighting the robustness provided by textual semantic guidance. This study offers an effective and scalable framework for few-shot segmentation with multimodal prompts.
- Open Access
- 6 January 2026
- Published by Elsevier in Expert Systems with Applications
- Vol. 308
BIG-PU: An evaluation metric for exploration based on preference elicitation in recommender systems
- 6 January 2026
- Published by Elsevier in Expert Systems with Applications
- Vol. 308
Open AccessBIG-PU: An evaluation metric for exploration based on preference elicitation in recommender systems
- Open Access
- 14 November 2025
- Published by MDPI AG in Applied Sciences
- Vol. 15 (22) , 12118
Continual Learning Model with Dynamic Adaptation and Flexibility for Incremental Classes
In recent years, deep learning models have achieved highly accurate image classification. However, in scenarios where the number of target classes continues to increase while using a model, it is not very effective to retrain the model by using all data from the beginning each time. Consequently, there has been increasing interest in class-incremental learning, which sequentially updates existing models with newly introduced classes. In particular, Forward Compatible Few-Shot Class-Incremental Learning (FACT) has demonstrated high classification performance by proactively reserving regions within the embedding space to represent knowledge of classes that will be added in the future. In this study, we propose FACT+, which seeks to further enhance the performance of FACT by incorporating two key extensions. First, to account for the semantic diversity (i.e., the breadth or narrowness of meaning) of each class, we introduce variability into each class’s covariance matrix. Second, to address the issue of insufficient learning from incremental data, we implement a mechanism to continuously update the model’s knowledge through the reuse of data collected during deployment. Experimental results show that the proposed method outperforms FACT in terms of classification accuracy and confirm the validity of the embedding space learned by the proposed approach.
- 14 November 2025
- Published by MDPI AG in Applied Sciences
- Vol. 15 (22) , 12118
Open AccessContinual Learning Model with Dynamic Adaptation and Flexibility for Incremental Classes
In recent years, deep learning models have achieved highly accurate image classification. However, in scenarios where the number of target classes continues to increase while using a model, it is not very effective to retrain the model by using all data from the beginning each time. Consequently, there has been increasing interest in class-incremental learning, which sequentially updates existing models with newly introduced classes. In particular, Forward Compatible Few-Shot Class-Incremental Learning (FACT) has demonstrated high classification performance by proactively reserving regions within the embedding space to represent knowledge of classes that will be added in the future. In this study, we propose FACT+, which seeks to further enhance the performance of FACT by incorporating two key extensions. First, to account for the semantic diversity (i.e., the breadth or narrowness of meaning) of each class, we introduce variability into each class’s covariance matrix. Second, to address the issue of insufficient learning from incremental data, we implement a mechanism to continuously update the model’s knowledge through the reuse of data collected during deployment. Experimental results show that the proposed method outperforms FACT in terms of classification accuracy and confirm the validity of the embedding space learned by the proposed approach.
- Open Access
- 2 September 2025
- Published by Elsevier in Neurocomputing
- Vol. 655
Generating realistic synthetic tabular data with integrated LLM and diffusion models
- 2 September 2025
- Published by Elsevier in Neurocomputing
- Vol. 655
Open AccessGenerating realistic synthetic tabular data with integrated LLM and diffusion models
- Open Access
- 1 September 2025
- Published by Elsevier in Knowledge-Based Systems
- Vol. 329
Optimizing pre-training via target-aware source data selection
- 1 September 2025
- Published by Elsevier in Knowledge-Based Systems
- Vol. 329
Open AccessOptimizing pre-training via target-aware source data selection
- Open Access
- 20 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 586-592
AI Ethics Education and Evaluation Framework Based on Case Method Using Network Motif and Sentiment Analysis
- 20 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 586-592
Open AccessAI Ethics Education and Evaluation Framework Based on Case Method Using Network Motif and Sentiment Analysis