In recent years, Large Language Models have revolutionized the field of natural language processing, showcasing an impressive rise predominantly in English-centric domains. These advancements have set a global benchmark, inspiring significant efforts toward developing Arabic LLMs capable of understanding and generating the Arabic language with remarkable accuracy. Despite these advancements, a critical challenge persists: the potential bias in Arabic LLMs, primarily attributed to their reliance on datasets comprising English data that has been translated into Arabic. This reliance not only compromises the authenticity of the generated content but also reflects a broader issue -the scarcity of original quality Arabic linguistic data. This study aims to address the data scarcity in the Arab world and to encourage the development of Arabic Language Models that are true to both the linguistic and nuances of the region. We undertook a large-scale data mining project, extracting a substantial volume of text from the Common Crawl WET files, specifically targeting Arabic content. The extracted data underwent a rigorous cleaning and deduplication process, using innovative techniques to ensure the integrity and uniqueness of the dataset. The result is the 101 Billion Arabic Words Dataset, the largest Arabic dataset available to date, which can significantly contribute to the development of authentic Arabic LLMs. This study not only highlights the potential for creating linguistically and culturally accurate Arabic LLMs but also sets a precedent for future research in enhancing the authenticity of Arabic language models.
Code translation tools (transpilers) are developed for automatic source-to-source translation. Although learning-based transpilers have shown impressive enhancement against rule-based counterparts, owing to their task-specific pre-training on extensive monolingual corpora. Their current performance still remains unsatisfactory for practical deployment, and the associated training resources are also prohibitively expensive. LLMs pre-trained on huge amounts of human-written code/text have shown remarkable performance in many code intelligence tasks due to their powerful generality, even without task-specific training. Thus, LLMs can potentially circumvent the above limitations, but they have not been exhaustively explored yet. This paper investigates diverse LLMs and learning-based transpilers for automated code translation tasks, finding that: although certain LLMs have outperformed current transpilers, they still have some accuracy issues, where most of the failures are induced by a lack of comprehension of source programs, missing clear instructions on I/O types in translation, and ignoring discrepancies between source and target programs. Enlightened by the above findings, we further propose UniTrans, a Unified code Translation framework, applicable to various LLMs, for unleashing their power in this field. Specifically, UniTrans first crafts a series of test cases for target programs with the assistance of source programs. Next, it harnesses the above auto-generated test cases to augment the code translation and then evaluate their correctness via execution. Afterward, UniTrans further (iteratively) repairs incorrectly translated programs prompted by test case execution results. Extensive experiments are conducted on six settings of translation datasets between Python, Java, and C++. Three recent LLMs of diverse sizes are tested with UniTrans, and all achieve substantial improvements.
Code translation between programming languages is a long-existing and critical task in software engineering, facilitating the modernization of legacy systems, ensuring cross-platform compatibility, and enhancing software performance. With the recent advances in large language models (LLMs) and their applications to code translation, there is an increasing need for comprehensive evaluation of these models. In this study, we empirically analyze the generated outputs of eleven popular instruct-tuned LLMs with parameters ranging from 1B up to 46.7B on 3,820 translation pairs across five languages, including C, C++, Go, Java, and Python. Our analysis found that between 26.4% and 73.7% of code translations produced by our evaluated LLMs necessitate post-processing, as these translations often include a mix of code, quotes, and text rather than being purely source code. Overlooking the output format of these models can inadvertently lead to underestimation of their actual performance. This is particularly evident when evaluating them with execution-based metrics such as Computational Accuracy (CA). Our results demonstrate that a strategic combination of prompt engineering and regular expression can effectively extract the source code from the model generation output. In particular, our method can help eleven selected models achieve an average Code Extraction Success Rate (CSR) of 92.73%. Our findings shed light on and motivate future research to conduct more reliable benchmarks of LLMs for code translation.
As Pre-trained Language Models (PLMs), a popular approach for code intelligence, continue to grow in size, the computational cost of their usage has become prohibitively expensive. Prompt learning, a recent development in the field of natural language processing, emerges as a potential solution to address this challenge. In this paper, we investigate the effectiveness of prompt learning in code intelligence tasks. We unveil its reliance on manually designed prompts, which often require significant human effort and expertise. Moreover, we discover existing automatic prompt design methods are very limited to code intelligence tasks due to factors including gradient dependence, high computational demands, and limited applicability. To effectively address both issues, we propose Genetic Auto Prompt (GenAP), which utilizes an elaborate genetic algorithm to automatically design prompts. With GenAP, non-experts can effortlessly generate superior prompts compared to meticulously manual-designed ones. GenAP operates without the need for gradients or additional computational costs, rendering it gradient-free and cost-effective. Moreover, GenAP supports both understanding and generation types of code intelligence tasks, exhibiting great applicability. We conduct GenAP on three popular code intelligence PLMs with three canonical code intelligence tasks including defect prediction, code summarization, and code translation. The results suggest that GenAP can effectively automate the process of designing prompts. Specifically, GenAP outperforms all other methods across all three tasks (e.g., improving accuracy by an average of 2.13% for defect prediction). To the best of our knowledge, GenAP is the first work to automatically design prompts for code intelligence PLMs.
Large language models (LLMs) such as ChatGPT are increasingly proficient in understanding and generating a mixture of code and text. Evaluation based on such $\textit{mixture}$ can lead to a more comprehensive understanding of the models' abilities in solving coding problems. However, in this context, current evaluation methods are either limited in task coverage or lack standardization. To address this issue, we propose using category theory as a framework for evaluation. Specifically, morphisms within a code category can represent code debugging and transformation, functors between two categories represent code translation, and functors between a code category and a natural language category represent code generation, explanation, and reproduction. We present an automatic evaluation framework called $\textbf{CatCode}$ ($\textbf{Cat}$egory $\textbf{Code}$) that can comprehensively assess the coding abilities of LLMs, including ChatGPT, Text-Davinci, and CodeGeeX.
Entity Alignment (EA) is vital for integrating diverse knowledge graph (KG) data, playing a crucial role in data-driven AI applications. Traditional EA methods primarily rely on comparing entity embeddings, but their effectiveness is constrained by the limited input KG data and the capabilities of the representation learning techniques. Against this backdrop, we introduce ChatEA, an innovative framework that incorporates large language models (LLMs) to improve EA. To address the constraints of limited input KG data, ChatEA introduces a KG-code translation module that translates KG structures into a format understandable by LLMs, thereby allowing LLMs to utilize their extensive background knowledge to improve EA accuracy. To overcome the over-reliance on entity embedding comparisons, ChatEA implements a two-stage EA strategy that capitalizes on LLMs' capability for multi-step reasoning in a dialogue format, thereby enhancing accuracy while preserving efficiency. Our experimental results verify ChatEA's superior performance, highlighting LLMs' potential in facilitating EA tasks.
Code generation models have increasingly become integral to aiding software development. Although current research has thoroughly examined the correctness of the code produced by code generation models, a vital aspect that plays a pivotal role in green computing and sustainability efforts has often been neglected. This paper presents EffiBench, a benchmark with 1,000 efficiency-critical coding problems to assess the efficiency of code generated by code generation models. EffiBench contains a diverse set of LeetCode coding problems. Each problem is paired with an executable human-written canonical solution, which obtains the SOTA efficiency on the LeetCode solution leaderboard. With EffiBench, we empirically examine the ability of 42 large language models (35 open-source and 7 closed-source) to generate efficient code. Our evaluation results demonstrate that the efficiency of the code generated by LLMs is generally worse than the efficiency of human-written canonical solutions. For example, GPT-4 generated code has an average \textbf{3.12} times execution time that of the human-written canonical solutions. In the most extreme cases, the execution time and total memory usage of GPT-4 generated code are \textbf{13.89} and \textbf{43.92} times that of the canonical solutions. The source code of EffiBench is released on https://github.com/huangd1999/EffiBench. We also provide the LeaderBoard at https://huggingface.co/spaces/EffiBench/effibench-leaderboard.
Code authorship attribution is the problem of identifying authors of programming language codes through the stylistic features in their codes, a topic that recently witnessed significant interest with outstanding performance. In this work, we present SCAE, a code authorship obfuscation technique that leverages a Seq2Seq code transformer called StructCoder. SCAE customizes StructCoder, a system designed initially for function-level code translation from one language to another (e.g., Java to C#), using transfer learning. SCAE improved the efficiency at a slight accuracy degradation compared to existing work. We also reduced the processing time by about 68% while maintaining an 85% transformation success rate and up to 95.77% evasion success rate in the untargeted setting.
Binary code analysis is the foundation of crucial tasks in the security domain; thus building effective binary analysis techniques is more important than ever. Large language models (LLMs) although have brought impressive improvement to source code tasks, do not directly generalize to assembly code due to the unique challenges of assembly: (1) the low information density of assembly and (2) the diverse optimizations in assembly code. To overcome these challenges, this work proposes a hierarchical attention mechanism that builds attention summaries to capture the semantics more effectively and designs contrastive learning objectives to train LLMs to learn assembly optimization. Equipped with these techniques, this work develops Nova, a generative LLM for assembly code. Nova outperforms existing techniques on binary code decompilation by up to 14.84 -- 21.58% (absolute percentage point improvement) higher Pass@1 and Pass@10, and outperforms the latest binary code similarity detection techniques by up to 6.17% Recall@1, showing promising abilities on both assembly generation and understanding tasks.
This paper introduces INTERVENOR (INTERactiVE chaiN Of Repair), a system designed to emulate the interactive code repair processes observed in humans, encompassing both code diagnosis and code repair. INTERVENOR prompts Large Language Models (LLMs) to play distinct roles during the code repair process, functioning as both a Code Learner and a Code Teacher. Specifically, the Code Learner is tasked with adhering to instructions to generate or repair code, while the Code Teacher is responsible for crafting a Chain-of-Repair (CoR) to serve as guidance for the Code Learner. During generating the CoR, the Code Teacher needs to check the generated codes from Code Learner and reassess how to address code bugs based on error feedback received from compilers. Experimental results demonstrate that INTERVENOR surpasses baseline models, exhibiting improvements of approximately 18% and 4.3% over GPT-3.5 in code generation and code translation tasks, respectively. Our further analyses show that CoR is effective to illuminate the reasons behind bugs and outline solution plans in natural language. With the feedback of code compilers, INTERVENOR can accurately identify syntax errors and assertion errors and provide precise instructions to repair codes. All data and codes are available at https://github.com/NEUIR/INTERVENOR
This work explores the use of self-generated natural language explanations as an intermediate step for code-to-code translation with language models. Across three types of explanations and 19 programming languages constructed from the MultiPL-E dataset, we find the explanations to be particularly effective in the zero-shot case, improving performance by 12% on average. Improvements with natural language explanations are particularly pronounced on difficult programs. We release our dataset, code, and canonical solutions in all 19 languages.
One major challenge of translating code between programming languages is that parallel training data is often limited. To overcome this challenge, we present two data augmentation techniques, one that builds comparable corpora (i.e., code pairs with similar functionality), and another that augments existing parallel data with multiple reference translations. Specifically, we build and analyze multiple types of comparable corpora, including programs generated from natural language documentation using a code generation model. Furthermore, to reduce overfitting to a single reference translation, we automatically generate additional translation references for available parallel data and filter the translations by unit tests, which increases variation in target translations. Experiments show that our data augmentation techniques significantly improve CodeT5 for translation between Java, Python, and C++ by an average of 7.5% Computational Accuracy (CA@1), which verifies the correctness of translations by execution. The code is available at https://github.com/Veronicium/CMTrans.
With the recent focus on Large Language Models (LLMs), both StarCoder (Li et al., 2023) and Code Llama (Rozi\`ere et al., 2023) have demonstrated remarkable performance in code generation. However, there is still a need for improvement in code translation functionality with efficient training techniques. In response to this, we introduce SteloCoder, a decoder-only StarCoder-based LLM designed specifically for multi-programming language-to-Python code translation. In particular, SteloCoder achieves C++, C#, JavaScript, Java, or PHP-to-Python code translation without specifying the input programming language. We modified StarCoder model architecture by incorporating a Mixture-of-Experts (MoE) technique featuring five experts and a gating network for multi-task handling. Experts are obtained by StarCoder fine-tuning. Specifically, we use a Low-Rank Adaptive Method (LoRA) technique, limiting each expert size as only 0.06% of number of StarCoder's parameters. At the same time, to enhance training efficiency in terms of time, we adopt curriculum learning strategy and use self-instruct data for efficient fine-tuning. As a result, each expert takes only 6 hours to train on one single 80Gb A100 HBM. With experiments on XLCoST datasets, SteloCoder achieves an average of 73.76 CodeBLEU score in multi-programming language-to-Python translation, surpassing the top performance from the leaderboard by at least 3.5. This accomplishment is attributed to only 45M extra parameters with StarCoder as the backbone and 32 hours of valid training on one 80GB A100 HBM. The source code is release here: https://github.com/sade-adrien/SteloCoder.
Code Large Language Models (Code LLMs) have gained significant attention in the industry due to their wide applications in the full lifecycle of software engineering. However, the effectiveness of existing models in understanding non-English inputs for multi-lingual code-related tasks is still far from well studied. This paper introduces CodeFuse-13B, an open-sourced pre-trained code LLM. It is specifically designed for code-related tasks with both English and Chinese prompts and supports over 40 programming languages. CodeFuse achieves its effectiveness by utilizing a high quality pre-training dataset that is carefully filtered by program analyzers and optimized during the training process. Extensive experiments are conducted using real-world usage scenarios, the industry-standard benchmark HumanEval-x, and the specially designed CodeFuseEval for Chinese prompts. To assess the effectiveness of CodeFuse, we actively collected valuable human feedback from the AntGroup's software development process where CodeFuse has been successfully deployed. The results demonstrate that CodeFuse-13B achieves a HumanEval pass@1 score of 37.10%, positioning it as one of the top multi-lingual code LLMs with similar parameter sizes. In practical scenarios, such as code generation, code translation, code comments, and testcase generation, CodeFuse performs better than other models when confronted with Chinese prompts.
Recent code translation techniques exploit neural machine translation models to translate source code from one programming language to another to satisfy production compatibility or to improve efficiency of codebase maintenance. Most existing code translation datasets only focus on a single pair of popular programming languages. To advance research on code translation and meet diverse requirements of real-world applications, we construct CodeTransOcean, a large-scale comprehensive benchmark that supports the largest variety of programming languages for code translation. CodeTransOcean consists of three novel multilingual datasets, namely, MultilingualTrans supporting translations between multiple popular programming languages, NicheTrans for translating between niche programming languages and popular ones, and LLMTrans for evaluating executability of translated code by large language models (LLMs). CodeTransOcean also includes a novel cross-framework dataset, DLTrans, for translating deep learning code across different frameworks. We develop multilingual modeling approaches for code translation and demonstrate their great potential in improving the translation quality of both low-resource and high-resource language pairs and boosting the training efficiency. We also propose a novel evaluation metric Debugging Success Rate@K for program-level code translation. Last but not least, we evaluate LLM ChatGPT on our datasets and investigate its potential for fuzzy execution predictions. We build baselines for CodeTransOcean and analyze challenges of code translation for guiding future research. The CodeTransOcean datasets and code are publicly available at https://github.com/WeixiangYAN/CodeTransOcean.
Large language models are becoming increasingly practical for translating code across programming languages, a process known as $transpiling$. Even though automated transpilation significantly boosts developer productivity, a key concern is whether the generated code is correct. Existing work initially used manually crafted test suites to test the translations of a small corpus of programs; these test suites were later automated. In contrast, we devise the first approach for automated, functional, property-based testing of code translation models. Our general, user-provided specifications about the transpiled code capture a range of properties, from purely syntactic to purely semantic ones. As shown by our experiments, this approach is very effective in detecting property violations in popular code translation models, and therefore, in evaluating model quality with respect to given properties. We also go a step further and explore the usage scenario where a user simply aims to obtain a correct translation of some code with respect to certain properties without necessarily being concerned about the overall quality of the model. To this purpose, we develop the first property-guided search procedure for code translation models, where a model is repeatedly queried with slightly different parameters to produce alternative and potentially more correct translations. Our results show that this search procedure helps to obtain significantly better code translations.
Neural Machine Translation (NMT) is widely applied in software engineering tasks. The effectiveness of NMT for code retrieval relies on the ability to learn from the sequence of tokens in the source language to the sequence of tokens in the target language. While NMT performs well in pseudocode-to-code translation, it might have challenges in learning to translate from natural language query to source code in newly curated real-world code documentation/ implementation datasets. In this work, we analyze the performance of NMT in natural language-to-code translation in the newly curated CAT benchmark that includes the optimized versions of three Java datasets TLCodeSum, CodeSearchNet, Funcom, and a Python dataset PCSD. Our evaluation shows that NMT has low accuracy, measured by CrystalBLEU and Meteor metrics in this task. To alleviate the duty of NMT in learning complex representation of source code, we propose ASTTrans Representation, a tailored representation of an Abstract Syntax Tree (AST) using a subset of non-terminal nodes. We show that the classical approach NMT performs significantly better in learning ASTTrans Representation over code tokens with up to 36% improvement on Meteor score. Moreover, we leverage ASTTrans Representation to conduct combined code search processes from the state-of-the-art code search processes using GraphCodeBERT and UniXcoder. Our NMT models of learning ASTTrans Representation can boost the Mean Reciprocal Rank of these state-of-the-art code search processes by up to 3.08% and improve 23.08% of queries' results over the CAT benchmark.
This paper presents a study on the feasibility of using large language models (LLM) for coding with low-resource and domain-specific programming languages that typically lack the amount of data required for effective LLM processing techniques. This study focuses on the econometric scripting language named hansl of the open-source software gretl and employs a proprietary LLM based on GPT-3.5. Our findings suggest that LLMs can be a useful tool for writing, understanding, improving, and documenting gretl code, which includes generating descriptive docstrings for functions and providing precise explanations for abstract and poorly documented econometric code. While the LLM showcased promoting docstring-to-code translation capability, we also identify some limitations, such as its inability to improve certain sections of code and to write accurate unit tests. This study is a step towards leveraging the power of LLMs to facilitate software development in low-resource programming languages and ultimately to lower barriers to entry for their adoption.
In this study, we present a novel dataset for training machine learning models translating between OpenMP Fortran and C++ code. To ensure reliability and applicability, the dataset is created from a range of representative open-source OpenMP benchmarks. It is also refined using a meticulous code similarity test. The effectiveness of our dataset is assessed using both quantitative (CodeBLEU) and qualitative (human evaluation) methods. We showcase how this dataset significantly elevates the translation competencies of large language models (LLMs). Specifically, models without prior coding knowledge experienced a boost of $\mathbf{\times~5.1}$ in their CodeBLEU scores, while models with some coding familiarity saw an impressive $\mathbf{\times~9.9}$-fold increase. The best fine-tuned model using our dataset outperforms GPT-4. It is also reaching human-level accuracy. This work underscores the immense potential of our dataset in propelling advancements in the domain of code translation for high-performance computing. The dataset is accessible at \href{https://github.com/bin123apple/Fortran-CPP-HPC-code-translation-dataset}{OpenMP-Fortran-CPP-Translation}.
This paper provides a comprehensive review of the literature concerning the utilization of Natural Language Processing (NLP) techniques, with a particular focus on transformer-based large language models (LLMs) trained using Big Code, within the domain of AI-assisted programming tasks. LLMs, augmented with software naturalness, have played a crucial role in facilitating AI-assisted programming applications, including code generation, code completion, code translation, code refinement, code summarization, defect detection, and clone detection. Notable examples of such applications include the GitHub Copilot powered by OpenAI's Codex and DeepMind AlphaCode. This paper presents an overview of the major LLMs and their applications in downstream tasks related to AI-assisted programming. Furthermore, it explores the challenges and opportunities associated with incorporating NLP techniques with software naturalness in these applications, with a discussion on extending AI-assisted programming capabilities to Apple's Xcode for mobile software development. This paper also presents the challenges of and opportunities for incorporating NLP techniques with software naturalness, empowering developers with advanced coding assistance and streamlining the software development process.