相关论文: Natural Language to Code Translation with Executio…

General-purpose LLM judges capable of human-level evaluation provide not only a scalable and accurate way of evaluating instruction-following LLMs but also new avenues for supervising and improving their performance. One promising way of…

Code translation is a crucial activity in the software development and maintenance process, and researchers have recently begun to focus on using pre-trained large language models (LLMs) for code translation. However, existing LLMs only…

Code-generating Large Language Models (LLMs) have become essential tools in modern software development, enhancing productivity and accelerating development. This paper aims to investigate the fine-tuning of code-generating LLMs using…

In this work, we propose minimum Bayes risk (MBR) training of RNN-Transducer (RNN-T) for end-to-end speech recognition. Specifically, initialized with a RNN-T trained model, MBR training is conducted via minimizing the expected edit…

To adequately test modern code generation systems, evaluation benchmarks must execute and test the code generated by the system. However, these execution and testing requirements have largely limited benchmarks to settings where code is…

Our goal is to learn a semantic parser that maps natural language utterances into executable programs when only indirect supervision is available: examples are labeled with the correct execution result, but not the program itself.…

Minimum Bayes-risk (MBR) decoding has recently gained renewed attention in text generation. MBR decoding considers texts sampled from a model as pseudo-references and selects the text with the highest similarity to the others. Therefore,…

Although large language models (LLMs) have been largely successful in generating functionally correct programs, conditioning models to produce efficient solutions while ensuring correctness remains a challenge. Further, unreliability in…

Evaluating the performance of Code Language Models (CLMs) for software engineering tasks, especially in multilingual and low-resource programming language settings, poses significant challenges. These challenges are primarily due to the…

One of the most important challenges in text generation systems is to produce outputs that are not only correct but also diverse. Recently, Minimum Bayes-Risk (MBR) decoding has gained prominence for generating sentences of the highest…

Program synthesis with language models (LMs) has unlocked a large set of reasoning abilities; code-tuned LMs have proven adept at generating programs that solve a wide variety of algorithmic symbolic manipulation tasks (e.g. word…

Current Text-to-Code models demonstrate impressive capabilities in generating executable code from natural language snippets. However, current studies focus on technical instructions and programmer-oriented language, and it is an open…

Detecting buffer overruns from a source code is one of the most common and yet challenging tasks in program analysis. Current approaches have mainly relied on rigid rules and handcrafted features devised by a few experts, limiting…

This paper proposes an adaptive neural-compilation framework to address the problem of efficient program learning. Traditional code optimisation strategies used in compilers are based on applying pre-specified set of transformations that…

Code generation models can benefit data scientists' productivity by automatically generating code from context and text descriptions. An important measure of the modeling progress is whether a model can generate code that can correctly…

Large language models (LLMs) have achieved notable success in code generation. However, they still frequently produce uncompilable output because their next-token inference procedure does not model formal aspects of code. Although…

Dynamically typed languages, like Erlang, allow developers to quickly write programs without explicitly providing any type information on expressions or function definitions. However, this feature makes those languages less reliable than…

Equivalent Representations (ERs) of code are textual representations that preserve the same semantics as the code itself, e.g., natural language comments and pseudocode. ERs play a critical role in software development and maintenance.…

Beam search is the most widely used decoding method for neural machine translation (NMT). In practice, the top-1 candidate with the highest log-probability among the n candidates is selected as the preferred one. However, this top-1…

Recent research has achieved impressive results on understanding and improving source code by building up on machine-learning techniques developed for natural languages. A significant advancement in natural-language understanding has come…