Cited by A Joint Learning Model with Variational Interaction for Multilingual Program Translation

Examining the Use and Impact of an AI Code Assistant on Developer Productivity and Experience in the Enterprise DOI

Justin D. Weisz, Shraddha Kumar, Michael Müller

et al.

Published: April 23, 2025

Language: Английский

Citations

Benchmarking large language models for automated labeling: The case of issue report classification DOI

Giuseppe Colavito, Filippo Lanubile, Nicole Novielli

et al.

Information and Software Technology, Journal Year: 2025, Volume and Issue: unknown, P. 107758 - 107758

Published: April 1, 2025

Language: Английский

Citations

Reproducibility in automated chemistry laboratories using computer science abstractions DOI

Richard B. Canty, Milad Abolhasani

Nature Synthesis, Journal Year: 2024, Volume and Issue: 3(11), P. 1327 - 1339

Published: Oct. 10, 2024

Language: Английский

Citations

Type-migrating C-to-Rust translation using a large language model DOI

Jaemin Hong, Sukyoung Ryu

Empirical Software Engineering, Journal Year: 2024, Volume and Issue: 30(1)

Published: Oct. 17, 2024

Abstract Rust, a modern system programming language, introduces new types that prevent memory bugs and data races. This makes translating legacy programs from C to Rust promising approach enhance their reliability. Since manual code translation is time-consuming, it desirable automate the translation. To yield satisfactory results, translator should have ability perform type migration , i.e., removing introducing in code. In this work, we aim automatically port an entire program by each function with signature containing proper through migration. goal challenging because (1) cannot be achieved syntactic mappings between names, (2) after migration, bodies correctly restructured based on precise understanding of functions’ behavior. address these difficulties, leverage large language models (LLMs), which possess knowledge semantics idioms. However, naïvely instructing LLMs translate produces unsatisfactory code, unmigrated or improperly migrated huge number errors. resolve issues, propose three techniques: generating candidate signatures, providing translated callees’ signatures LLMs, (3) iteratively fixing errors using compiler feedback. Our evaluation shows proposed yields 63.5% increase 71.5% decrease compared baseline (the naïve LLM-based translation) modest performance overhead.

Language: Английский

Citations

Rust for Embedded Systems: Current State and Open Problems DOI

Ayushi Sharma, Shashank Sharma, Sai Ritvik Tanksalkar

et al.

Published: Dec. 2, 2024

Embedded software is used in safety-critical systems such as medical devices and autonomous vehicles, where defects, including security vulnerabilities, have severe consequences. Most embedded codebases are developed unsafe languages, specifically C/C++, riddled with memory safety vulnerabilities. To prevent Rust, a performant memory-safe language, provides an optimal choice for developing software. Rust interoperability enables applications on top of existing C codebases. Despite this, even the most resourceful organizations continue to develop C/C++.This paper performs first systematic study holistically understand current state challenges using systems. Our organized across three research questions. We collected dataset 6,408 spanning various categories 6 Static Application Security Testing (SAST) tools. performed analysis our surveys 225 developers investigate found that support inadequate, SAST tools cannot handle certain features software, resulting failures, prevalence advanced types makes it challenging engineer interoperable code. In addition, we faced by development.

Language: Английский

Citations

SimAC: simulating agile collaboration to generate acceptance criteria in user story elaboration DOI