CodeFill

Code completion has become an essential component of integrated development environments. Contemporary code completion methods rely on the abstract syntax tree (AST) to generate syntactically correct code. However, they cannot fully capture the sequential and repetitive patterns of writing code and the structural information of the AST. To alleviate these problems, we propose a new code completion approach named CCAG, which models the flattened sequence of a partial AST as an AST graph. CCAG uses our proposed AST Graph Attention Block to capture different dependencies in the AST graph for representation learning in code completion. The sub-tasks of code completion are optimized via multi-task learning in CCAG, and the task balance is automatically achieved using uncertainty without the need to tune task weights. The experimental results show that CCAG has superior performance than state-of-the-art approaches and it is able to provide intelligent code completion.

Code completion, one of the most useful features in the Integrated Development Environments (IDEs), can accelerate software development by suggesting the libraries, APIs, and method names in real-time. Recent studies have shown that statistical language models can improve the performance of code completion tools through learning from large-scale software repositories. However, these models suffer from three major drawbacks: a) The hierarchical structural information of the programs is not fully utilized in the program's representation; b) In programs, the semantic relationships can be very long. Existing recurrent neural networks based language models are not sufficient to model the long-term dependency. c) Existing approaches perform a specific task in one model, which leads to the underuse of the information from related tasks. To address these challenges, in this paper, we propose a self-attentional neural architecture for code completion with multi-task learning. To utilize the hierarchical structural information of the programs, we present a novel method that considers the path from the predicting node to the root node. To capture the long-term dependency in the input programs, we adopt a self-attentional architecture based network as the base language model. To enable the knowledge sharing between related tasks, we creatively propose a Multi-Task Learning (MTL) framework to learn two related tasks in code completion jointly. Experiments on three real-world datasets demonstrate the effectiveness of our model when compared with state-of-the-art methods.

Traditional program slicing techniques are crucial for early bug detection and manual/automated debugging of online code snippets. Nevertheless, their inability to handle incomplete code hinders their real-world applicability in such scenarios. To overcome these challenges, we present NS-Slicer, a novel learning-based approach that predicts static program slices for both complete and partial code Our tool leverages a pre-trained language model to exploit its understanding of fine-grained variable-statement dependencies within source code. With this knowledge, given a variable at a specific location and a statement in a code snippet, NS-Slicer determines whether the statement belongs to the backward slice or forward slice, respectively. We conducted a series of experiments to evaluate NS-Slicer's performance. On complete code, it predicts the backward and forward slices with an F1-score of 97.41% and 95.82%, respectively, while achieving an overall F1-score of 96.77%. Notably, in 85.20% of the cases, the static program slices predicted by NS-Slicer exactly match entire slices from the oracle. For partial programs, it achieved an F1-score of 96.77%–97.49% for backward slicing, 92.14%–95.40% for forward slicing, and an overall F1-score of 94.66%–96.62%. Furthermore, we demonstrate NS-Slicer's utility in vulnerability detection (VD), integrating its predicted slices into an automated VD tool. In this setup, the tool detected vulnerabilities in Java code with a high F1-score of 73.38%. We also include the analyses studying NS-Slicer’s promising performance and limitations, providing insights into its understanding of intrinsic code properties such as variable aliasing, leading to better slicing.

Code completion is one of the most useful features in the Integrated Development Environments (IDEs), which can accelerate software development by suggesting the next probable token based on the contextual code in real-time. Recent studies have shown that statistical language modeling techniques can improve the performance of code completion tools through learning from large-scale software repositories. However, these models suffer from two major drawbacks: a) Existing research uses static embeddings, which map a word to the same vector regardless of its context. The differences in the meaning of a token in varying contexts are lost when each token is associated with a single representation; b) Existing language model based code completion models perform poor on completing identifiers, and the type information of the identifiers is ignored in most of these models. To address these challenges, in this paper, we develop a multi-task learning based pre-trained language model for code understanding and code generation with a Transformer-based neural architecture. We pre-train it with hybrid objective functions that incorporate both code understanding and code generation tasks. Then we fine-tune the pre-trained model on code completion. During the completion, our model does not directly predict the next token. Instead, we adopt multi-task learning to predict the token and its type jointly and utilize the predicted type to assist the token prediction. Experiments results on two real-world datasets demonstrate the effectiveness of our model when compared with state-of-the-art methods.

The Ames mutagenicity test constitutes the most frequently used assay to estimate the mutagenic potential of drug candidates. While this test employs experimental results using various strains of Salmonella typhimurium, the vast majority of the published in silico models for predicting mutagenicity do not take into account the test results of the individual experiments conducted for each strain. Instead, such QSAR models are generally trained employing overall labels (i.e., mutagenic and nonmutagenic). Recently, neural-based models combined with multitask learning strategies have yielded interesting results in different domains, given their capabilities to model multitarget functions. In this scenario, we propose a novel neural-based QSAR model to predict mutagenicity that leverages experimental results from different strains involved in the Ames test by means of a multitask learning approach. To the best of our knowledge, the modeling strategy hereby proposed has not been applied to model Ames mutagenicity previously. The results yielded by our model surpass those obtained by single-task modeling strategies, such as models that predict the overall Ames label or ensemble models built from individual strains. For reproducibility and accessibility purposes, all source code and datasets used in our experiments are publicly available.

Specialized Multi-Task Optimizers (SMTOs) balance task learning in Multi-Task Learning by addressing issues like conflicting gradients and differing gradient norms, which hinder equal-weighted task training. However, recent critiques suggest that equally weighted tasks can achieve competitive results compared to SMTOs, arguing that previous SMTO results were influenced by poor hyperparameter optimization and lack of regularization. In this work, we evaluate these claims through an extensive empirical evaluation of SMTOs, including some of the latest methods, on more complex multi-task problems to clarify this behavior. More specifically, we start our analysis by evaluating all SMTOs on a simple MNIST problem to identify the promising optimizers and evaluate then further on progressively more complex Multi-Task problems. Our findings indicate that SMTOs perform well compared to uniform loss and that fixed weights can achieve competitive performance compared to SMTOs. Furthermore, we demonstrate why uniform loss perform similarly to SMTOs in some instances. The source code is available at https://github.com/Gabriel-SGama/UnitScal_vs_SMTOs.

Chinese Event Detection (CED) aims to detect events from unstructured sentences. Due to the difficulty of labeling event detection datasets, previous approaches suffer from severe data sparsity problem. To address this issue, we propose a novel Lattice LSTM based multi-task learning model. On one hand, we utilize multi-granularity word information via Lattice LSTM to fully exploit existing datasets. On the other hand, we employ the multi-task learning mechanism to improve CED with datasets from other tasks. Specifically, we combine Name Entity Recognition (NER) and Mask Word Prediction (MWP) as two auxiliary tasks to learn both entity and general language information. Experiments show that our approach outperforms the six SOTA methods by 1.9% on ACE2005 benchmark. The source code is released on https://github.com/tongmeihan1995/MLL-chinese-event-detection.

An IDE code completion mechanism for stepwise API exploration.A method for building a language model for supporting (1).a detailed evaluation of (2), showing how n-grams perform against existing client code.Results of (3) revealing a 90% chance of getting the expected result in the 10-top-most code completion proposals. Reusing software involves learning third-party APIs, a process that is often time-consuming and error-prone. Recommendation systems for API usage assistance based on statistical models built from source code corpora are capable of assisting API users through code completion mechanisms in IDEs. A valid sequence of API calls involving different types may be regarded as a well-formed sentence of tokens from the API vocabulary. In this article we describe an approach for recommending subsequent tokens to complete API sentences using n-gram language models built from source code corpora. The provided system was integrated in the code completion facilities of the Eclipse IDE, providing contextualized completion proposals for Java taking into account the nearest lines of code. The approach was evaluated against existing client code of four widely used APIs, revealing that in more than 90% of the cases the expected subsequent token is within the 10-top-most proposals of our models. The high score provides evidence that the recommendations could help on API learning and exploration, namely through the assistance on writing valid API sentences.

Context: A large number of source code metrics are reported in the literature. It is necessary to systematically collect, describe and classify source code metrics to support research and practice. Objective: We aim to utilize existing secondary studies to develop a catalog of source code metrics together with their descriptions. The catalog will also provide information about which units of code (e.g., operators, operands, lines of code, variables, parameters, code blocks, or functions) are used to measure the internal quality attributes and the scope on which they are collected. Method: We conducted a tertiary study to identify secondary studies reporting source code metrics. We have classified the source code metrics according to the measured internal quality attributes, the units of code used in the measures, and the scope at which the source code metrics are collected. Results: From 711 secondary studies, we identified 52 relevant secondary studies. We reported 423 source code metrics together with their descriptions and the internal quality attributes they measure. Source code metrics predominantly incorporate function as a unit of code to measure internal quality attributes. In contrast, several source code metrics use more than one unit of code when measuring internal quality attributes. Nearly 51% of the source code metrics are collected at the class scope, while almost 12% and 15% of source code metrics are collected at module and application levels, respectively. Conclusions: Researchers and practitioners can use the extensive catalog to assess which source code metrics meet their individual needs based on the description and classification scheme presented.

National classifications and terminologies already routinely used for documentation within patient care settings enable the unambiguous representation of clinical information. However, the diversity of different vocabularies across health care institutions and countries is a barrier to achieving semantic interoperability and exchanging data across sites. The Observational Medical Outcomes Partnership (OMOP) Common Data Model (CDM) enables the standardization of structure and medical terminology. It allows the mapping of national vocabularies into so-called standard concepts, representing normative expressions for international analyses and research. Within our project "Hybrid Quality Indicators Using Machine Learning Methods" (Hybrid-QI), we aim to harmonize source codes used in German claims data vocabularies that are currently unavailable in the OMOP CDM. This study aims to increase the coverage of German vocabularies in the OMOP CDM. We aim to completely transform the source codes used in German claims data into the OMOP CDM without data loss and make German claims data usable for OMOP CDM-based research. To prepare the missing German vocabularies for the OMOP CDM, we defined a vocabulary preparation approach consisting of the identification of all codes of the corresponding vocabularies, their assembly into machine-readable tables, and the translation of German designations into English. Furthermore, we used 2 proposed approaches for OMOP-compliant vocabulary preparation: the mapping to standard concepts using the Observational Health Data Sciences and Informatics (OHDSI) tool Usagi and the preparation of new 2-billion concepts (ie, concept_id >2 billion). Finally, we evaluated the prepared vocabularies regarding completeness and correctness using synthetic German claims data and calculated the coverage of German claims data vocabularies in the OMOP CDM. Our vocabulary preparation approach was able to map 3 missing German vocabularies to standard concepts and prepare 8 vocabularies as new 2-billion concepts. The completeness evaluation showed that the prepared vocabularies cover 44.3% (3288/7417) of the source codes contained in German claims data. The correctness evaluation revealed that the specified validity periods in the OMOP CDM are compliant for the majority (705,531/706,032, 99.9%) of source codes and associated dates in German claims data. The calculation of the vocabulary coverage showed a noticeable decrease of missing vocabularies from 55% (11/20) to 10% (2/20) due to our preparation approach. By preparing 10 vocabularies, we showed that our approach is applicable to any type of vocabulary used in a source data set. The prepared vocabularies are currently limited to German vocabularies, which can only be used in national OMOP CDM research projects, because the mapping of new 2-billion concepts to standard concepts is missing. To participate in international OHDSI network studies with German claims data, future work is required to map the prepared 2-billion concepts to standard concepts.

Code summarization generates brief natural language description given a source code snippet, while code retrieval fetches relevant source code given a natural language query. Since both tasks aim to model the association between natural language and programming language, recent studies have combined these two tasks to improve their performance. However, researchers have yet been able to effectively leverage the intrinsic connection between the two tasks as they train these tasks in a separate or pipeline manner, which means their performance can not be well balanced. In this paper, we propose a novel end-to-end model for the two tasks by introducing an additional code generation task. More specifically, we explicitly exploit the probabilistic correlation between code summarization and code generation with dual learning, and utilize the two encoders for code summarization and code generation to train the code retrieval task via multi-task learning. We have carried out extensive experiments on an existing dataset of SQL and Python, and results show that our model can significantly improve the results of the code retrieval task over the-state-of-art models, as well as achieve competitive performance in terms of BLEU score for the code summarization task.

Multi-source binary translation system (MBT) provides a multi-source platform code to a single target platform migration solution. Do not know the source code, you can achieve the multi-source platform, multi-compiled version of the executable file on another platform. In order to achieve the special semantic mapping function of the MBT system, the need for a unified, compatible with multiple target platform differences, atomic semantic description language. In this paper, we design an atomic semantic description language MSI-SDL (multi-source instruction semantic description language), the x86-32, x86-64, power platform code, to class alpha platform architecture code semantic mapping function in the migration process. MBT base on MSI-SDL are evaluated wholly and partIy across the SPEC 2006 under the Alpha multi-core environment, test results verify the MSI-SDL correctness and validity.

A simple and compact Python code for complex 3D topology optimization

Multi-core shared memory architectures have become ubiquitous in computing hardware nowadays. As a result, there is a growing need to fully utilize these architectures by introducing appropriate parallelization schemes, such as OpenMP worksharing-loop constructs, to applications. However, most developers find introducing OpenMP directives to their code hard due to pervasive pitfalls in managing parallel shared memory. To assist developers in this process, many compilers, as well as source-to-source (S2S) translation tools, have been developed over the years, tasked with inserting OpenMP directives into code automatically. In addition to having limited robustness to their input format, these compilers still do not achieve satisfactory coverage and precision in locating parallelizable code and generating appropriate directives. Recently, many data-driven AI-based code completion (CC) tools, such as GitHub CoPilot, have been developed to ease and improve programming productivity. Leveraging the insights from existing AI-based programming-assistance tools, this work presents a novel AI model that can serve as a parallel-programming assistant. Specifically, our model, named PragFormer, is tasked with identifying for loops that can benefit from conversion to parallel worksharing-loop construct (OpenMP directive) and even predict the need for specific data-sharing attributes clauses on the fly. We created a unique database, named Open-OMP, specifically for this goal. Open-OMP contains over 32,000 unique code snippets from different domains, half of which contain OpenMP directives, while the other half do not. We experimented with different model design parameters for these tasks and showed that our best-performing model outperforms a statistically-trained baseline as well as a state-of-the-art S2S compiler. In fact, it even outperforms the popular generative AI model of ChatGPT. In the spirit of advancing research on this topic, we have already released source code for PragFormer as well as Open-OMP dataset to public. Moreover, an interactive demo of our tool, as well as a Hugging Face webpage to experiment with our tool, are already available.

Studies have confirmed the robust performance of machine learning classifiers for various source code modeling tasks. In general, machine learning approaches are incapable of handling imbalanced datasets, since they are sensitive to the choice of diverse classes. Therefore, these approaches may lean towards the classes with a large percentage of observations. In this work, we investigate and explore the impact of balanced and imbalanced learning on source code suggestion task otherwise known as code completion, covering a large number of imbalanced classes. We further explore the impact of vocabulary size on modeling performance. First, we provide the essentials to formulate the problem of source code suggestion as a classification task and investigate the level of imbalanced classes. Second, we train the four most adapted neural language models as a baseline to assess the modeling performance. Third, we impose two diverse class balancing techniques, TomekLinks and AllKNN, to balance the datasets and evaluate their impact on the modeling performance. Finally, we trained these models with a weighted imbalanced learning approach and compared the performance with balanced learning approaches. Additionally, we train models by varying the vocabulary size to study their impact. In total, we trained 230 models on 10 real-world software projects and extensively evaluated these models with widely used performance metrics such as Precision, Recall, FScore, mean reciprocal rank (MRR), and Receiver operating characteristics (ROC). Additionally, we employed ANOVA statistical analysis to study the statistical significance and differences between these approaches. This study has demonstrated that the modeling performance decreases during balanced model training, whereas the weighted imbalance training produces comparable results and is more efficient in terms of time cost. Additionally, this study exhibits that a large size of vocabulary does not necessarily improve the modeling performance when out-of-vocabulary predictions are disregarded.