Yutaka Watanobe

Atsushi Shirafuji, Yusuke Oda, Jun Suzuki et al.

A less complex and more straightforward program is a crucial factor that enhances its maintainability and makes writing secure and bug-free programs easier. However, due to its heavy workload and the risks of breaking the working programs, programmers are reluctant to do code refactoring, and thus, it also causes the loss of potential learning experiences. To mitigate this, we demonstrate the application of using a large language model (LLM), GPT-3.5, to suggest less complex versions of the user-written Python program, aiming to encourage users to learn how to write better programs. We propose a method to leverage the prompting with few-shot examples of the LLM by selecting the best-suited code refactoring examples for each target programming problem based on the prior evaluation of prompting with the one-shot example. The quantitative evaluation shows that 95.68% of programs can be refactored by generating 10 candidates each, resulting in a 17.35% reduction in the average cyclomatic complexity and a 25.84% decrease in the average number of lines after filtering only generated programs that are semantically correct. Furthermore, the qualitative evaluation shows outstanding capability in code formatting, while unnecessary behaviors such as deleting or translating comments are also observed.

Atsushi Shirafuji, Yutaka Watanobe, Takumi Ito et al.

Using large language models (LLMs) for source code has recently gained attention. LLMs, such as Transformer-based models like Codex and ChatGPT, have been shown to be highly capable of solving a wide range of programming problems. However, the extent to which LLMs understand problem descriptions and generate programs accordingly or just retrieve source code from the most relevant problem in training data based on superficial cues has not been discovered yet. To explore this research question, we conduct experiments to understand the robustness of several popular LLMs, CodeGen and GPT-3.5 series models, capable of tackling code generation tasks in introductory programming problems. Our experimental results show that CodeGen and Codex are sensitive to the superficial modifications of problem descriptions and significantly impact code generation performance. Furthermore, we observe that Codex relies on variable names, as randomized variables decrease the solved rate significantly. However, the state-of-the-art (SOTA) models, such as InstructGPT and ChatGPT, show higher robustness to superficial modifications and have an outstanding capability for solving programming problems. This highlights the fact that slight modifications to the prompts given to the LLMs can greatly affect code generation performance, and careful formatting of prompts is essential for high-quality code generation, while the SOTA models are becoming more robust to perturbations.

Atsushi Shirafuji, Md. Mostafizer Rahman, Md Faizul Ibne Amin et al.

Programmers often struggle to identify and fix bugs in their programs. In recent years, many language models (LMs) have been proposed to fix erroneous programs and support error recovery. However, the LMs tend to generate solutions that differ from the original input programs. This leads to potential comprehension difficulties for users. In this paper, we propose an approach to suggest a correct program with minimal repair edits using CodeT5. We fine-tune a pre-trained CodeT5 on code pairs of wrong and correct programs and evaluate its performance with several baseline models. The experimental results show that the fine-tuned CodeT5 achieves a pass@100 of 91.95% and an average edit distance of the most similar correct program of 6.84, which indicates that at least one correct program can be suggested by generating 100 candidate programs. We demonstrate the effectiveness of LMs in suggesting program repair with minimal edits for solving introductory programming problems.

Atsushi Shirafuji, Yutaka Watanobe

Referring to solution programs written by other users is helpful for learners in programming education. However, current online judge systems just list all solution programs submitted by users for references, and the programs are sorted based on the submission date and time, execution time, or user rating, ignoring to what extent the programs can be helpful to be referenced. In addition, users struggle to refer to a variety of solution approaches since there are too many duplicated and near-duplicated programs. To motivate learners to refer to various solutions to learn better solution approaches, in this paper, we propose an approach to deduplicate and rank common solution programs in each programming problem. Inspired by the nature that the many-duplicated program adopts a more common approach and can be a general reference, we remove the near-duplicated solution programs and rank the unique programs based on the duplicate count. The experiments on the solution programs submitted to a real-world online judge system demonstrate that the number of programs is reduced by 60.20%, whereas the baseline only reduces by 29.59% after the deduplication, meaning that users only need to refer to 39.80% of programs on average. Furthermore, our analysis shows that top-10 ranked programs cover 29.95% of programs on average, indicating that users can grasp 29.95% of solution approaches by referring to only 10 programs. The proposed approach shows the potential of reducing the learners' burden of referring to too many solutions and motivating them to learn a variety of solution approaches.

Md Mostafizer Rahman, Ariful Islam Shiplu, Yutaka Watanobe et al.

Contextual embeddings generated by LLMs exhibit strong positional inductive biases, which can limit their ability to fully capture long-range, order-sensitive dependencies in highly structured source code. Consequently, how to further refine and enhance LLM embeddings for improved code understanding remains an open research question. To address this gap, we propose a hybrid LLM-RNN framework that reinforces LLM-generated contextual embeddings with a sequential RNN architecture. The embeddings reprocessing step aims to reinforce sequential semantics and strengthen order-aware dependencies inherent in source code. We evaluate the proposed hybrid models on both benchmark and real-world coding datasets. The experimental results show that the RoBERTa-BiGRU and CodeBERT-GRU models achieved accuracies of 66.40% and 66.03%, respectively, on the defect detection benchmark dataset, representing improvements of approximately 5.35% and 3.95% over the standalone RoBERTa and CodeBERT models. Furthermore, the CodeT5-GRU and CodeT5+-BiGRU models achieved accuracies of 67.90% and 67.79%, respectively, surpassing their base models and outperforming RoBERTa-BiGRU and CodeBERT-GRU by a notable margin. In addition, CodeT5-GRU model attains weighted and macro F1-scores of 67.18% and 67.00%, respectively, on the same dataset. Extensive experiments across three real-world datasets further demonstrate consistent and statistically significant improvements over standalone LLMs. Overall, our findings indicate that reprocessing contextual embeddings with RNN architectures enhances code understanding performance in LLM-based models.

Md Mostafizer Rahman, Ariful Islam Shiplu, Md Faizul Ibne Amin et al.

Pretrained Large Language Models (LLMs) have achieved remarkable success across diverse domains, with education and research emerging as particularly impactful areas. Among current state-of-the-art LLMs, ChatGPT and DeepSeek exhibit strong capabilities in mathematics, science, medicine, literature, and programming. In this study, we present a comprehensive evaluation of these two LLMs through background technology analysis, empirical experiments, and a real-world user survey. The evaluation explores trade-offs among model accuracy, computational efficiency, and user experience in educational and research affairs. We benchmarked these LLMs performance in text generation, programming, and specialized problem-solving. Experimental results show that ChatGPT excels in general language understanding and text generation, while DeepSeek demonstrates superior performance in programming tasks due to its efficiency-focused design. Moreover, both models deliver medically accurate diagnostic outputs and effectively solve complex mathematical problems. Complementing these quantitative findings, a survey of students, educators, and researchers highlights the practical benefits and limitations of these models, offering deeper insights into their role in advancing education and research.

Atsushi Shirafuji, Taku Matsumoto, Md Faizul Ibne Amin et al.

Finding and fixing errors is a time-consuming task not only for novice programmers but also for expert programmers. Prior work has identified frequent error patterns among various levels of programmers. However, the differences in the tendencies between novices and experts have yet to be revealed. From the knowledge of the frequent errors in each level of programmers, instructors will be able to provide helpful advice for each level of learners. In this paper, we propose a rule-based error classification tool to classify errors in code pairs consisting of wrong and correct programs. We classify errors for 95,631 code pairs and identify 3.47 errors on average, which are submitted by various levels of programmers on an online judge system. The classified errors are used to analyze the differences in frequent errors between novice and expert programmers. The analyzed results show that, as for the same introductory problems, errors made by novices are due to the lack of knowledge in programming, and the mistakes are considered an essential part of the learning process. On the other hand, errors made by experts are due to misunderstandings caused by the carelessness of reading problems or the challenges of solving problems differently than usual. The proposed tool can be used to create error-labeled datasets and for further code-related educational research.

Md Faizul Ibne Amin, Yutaka Watanobe, Md. Mostafizer Rahman et al.

Programming is a core skill in computer science and software engineering (SE), yet identifying and resolving code errors remains challenging for both novice and experienced developers. While Large Language Models (LLMs) have shown remarkable capabilities in natural language understanding and generation tasks, their potential in domain-specific, complex scenarios, such as multi-label classification (MLC) of programming errors, remains underexplored. Recognizing this less-explored area, this study proposes a multi-label error classification (MLEC) framework for source code that leverages fine-tuned LLMs, including CodeT5-base, GraphCodeBERT, CodeT5+, UniXcoder, RoBERTa, PLBART, and CoTexT. These LLMs are integrated with deep learning (DL) architectures such as GRU, LSTM, BiLSTM, and BiLSTM with an additive attention mechanism (BiLSTM-A) to capture both syntactic and semantic features from a real-world student-written Python code error dataset. Extensive experiments across 32 model variants, optimized using Optuna-based hyperparameter tuning, have been evaluated using comprehensive multi-label metrics, including average accuracy, macro and weighted precision, recall, F1-score, exact match accuracy, One-error, Hamming loss, Jaccard similarity, and ROC-AUC (micro, macro, and weighted). Results show that the CodeT5+\_GRU model achieved the strongest performance, with a weighted F1-score of 0.8243, average accuracy of 91.84\%, exact match accuracy of 53.78\%, Hamming loss of 0.0816, and One error of 0.0708. These findings confirm the effectiveness of combining pretrained semantic encoders with efficient recurrent decoders. This work lays the foundation for developing intelligent, scalable tools for automated code feedback, with potential applications in programming education (PE) and broader SE domains.

Md Abrar Jahin, Md Sakib Hossain Shovon, M. F. Mridha et al.

Sentiment analysis is crucial for understanding public opinion and consumer behavior. Existing models face challenges with linguistic diversity, generalizability, and explainability. We propose TRABSA, a hybrid framework integrating transformer-based architectures, attention mechanisms, and BiLSTM networks to address this. Leveraging RoBERTa-trained on 124M tweets, we bridge gaps in sentiment analysis benchmarks, ensuring state-of-the-art accuracy. Augmenting datasets with tweets from 32 countries and US states, we compare six word-embedding techniques and three lexicon-based labeling techniques, selecting the best for optimal sentiment analysis. TRABSA outperforms traditional ML and deep learning models with 94% accuracy and significant precision, recall, and F1-score gains. Evaluation across diverse datasets demonstrates consistent superiority and generalizability. SHAP and LIME analyses enhance interpretability, improving confidence in predictions. Our study facilitates pandemic resource management, aiding resource planning, policy formation, and vaccination tactics.

Md Faizul Ibne Amin, Yutaka Watanobe, Daniel M. Muepu et al.

LLMs are increasingly employed both as judges for evaluating open-ended outputs and as co-creation partners in AI-assisted programming; yet rigorous evaluation in human-AI co-creation settings remains underdeveloped as judgments must be reliable, comparable across models, and interpretable over multi-turn interaction. To address this gap, a rubric-driven LLM-as-a-Judge framework is presented for contest-style human-AI co-creation in coding and software engineering (SE). The framework is built around schema-constrained judge outputs, validation and repair mechanisms, grouped and split by user and problem to prevent trajectory leakage, and participant-level NONBLIND context. Multiple LLM judges are assessed through a multi-metric protocol covering discrimination (ROC-AUC, PR-AUC), thresholded decision quality (MCC), probabilistic reliability (LogLoss, Brier score, ECE), and inter-judge agreement (Cohen's and Fleiss' k). Human-AI co-creation is further examined through trajectory-level signals, including turn-wise confidence, Success-at-Turn, time-to-success, revision churn, and CodeBLEU. Co-creation success is found to concentrate early, with Success-at-Turn rising to 0.8533 at the first observed turn and stabilizing at 0.8641 by turn 6. Revision behavior, however, remains heterogeneous, suggesting that productive progress can emerge through either incremental refinement or broader restructuring. On the judging side, the best held-out scores reach 0.5937 for ROC-AUC, 0.6904 for PR-AUC, and 0.5000 for MCC test, while inter-judge consistency remains modest overall (mean pairwise Cohen's k = 0.1592, Fleiss' k = 0.0696). Taken together, this work offers an auditable and reproducible evaluation methodology that links reliability-aware LLM judging with trajectory-based analysis of human-AI co-creation, providing a practical evaluation template for future AI-assisted coding and SE.

Md. Mostafizer Rahman, Ariful Islam Shiplu, Yutaka Watanobe et al.

Effectively analyzing the comments to uncover latent intentions holds immense value in making strategic decisions across various domains. However, several challenges hinder the process of sentiment analysis including the lexical diversity exhibited in comments, the presence of long dependencies within the text, encountering unknown symbols and words, and dealing with imbalanced datasets. Moreover, existing sentiment analysis tasks mostly leveraged sequential models to encode the long dependent texts and it requires longer execution time as it processes the text sequentially. In contrast, the Transformer requires less execution time due to its parallel processing nature. In this work, we introduce a novel hybrid deep learning model, RoBERTa-BiLSTM, which combines the Robustly Optimized BERT Pretraining Approach (RoBERTa) with Bidirectional Long Short-Term Memory (BiLSTM) networks. RoBERTa is utilized to generate meaningful word embedding vectors, while BiLSTM effectively captures the contextual semantics of long-dependent texts. The RoBERTa-BiLSTM hybrid model leverages the strengths of both sequential and Transformer models to enhance performance in sentiment analysis. We conducted experiments using datasets from IMDb, Twitter US Airline, and Sentiment140 to evaluate the proposed model against existing state-of-the-art methods. Our experimental findings demonstrate that the RoBERTa-BiLSTM model surpasses baseline models (e.g., BERT, RoBERTa-base, RoBERTa-GRU, and RoBERTa-LSTM), achieving accuracies of 80.74%, 92.36%, and 82.25% on the Twitter US Airline, IMDb, and Sentiment140 datasets, respectively. Additionally, the model achieves F1-scores of 80.73%, 92.35%, and 82.25% on the same datasets, respectively.

M. F. Mridha, Abu Quwsar Ohi, Muhammad Mostafa Monowar et al.

Speaker recognition deals with recognizing speakers by their speech. Most speaker recognition systems are built upon two stages, the first stage extracts low dimensional correlation embeddings from speech, and the second performs the classification task. The robustness of a speaker recognition system mainly depends on the extraction process of speech embeddings, which are primarily pre-trained on a large-scale dataset. As the embedding systems are pre-trained, the performance of speaker recognition models greatly depends on domain adaptation policy, which may reduce if trained using inadequate data. This paper introduces a speaker recognition strategy dealing with unlabeled data, which generates clusterable embedding vectors from small fixed-size speech frames. The unsupervised training strategy involves an assumption that a small speech segment should include a single speaker. Depending on such a belief, a pairwise constraint is constructed with noise augmentation policies, used to train AutoEmbedder architecture that generates speaker embeddings. Without relying on domain adaption policy, the process unsupervisely produces clusterable speaker embeddings, termed unsupervised vectors (u-vectors). The evaluation is concluded in two popular speaker recognition datasets for English language, TIMIT, and LibriSpeech. Also, a Bengali dataset is included to illustrate the diversity of the domain shifts for speaker recognition systems. Finally, we conclude that the proposed approach achieves satisfactory performance using pairwise architectures.