paper-with-me

Papers

PanGu-Coder: Program Synthesis with Function-Level Language Modeling

2022-07-22 · Fenia Christopoulou, Gerasimos Lampouras, Milan Gritta, Guchun Zhang, Yinpeng Guo, Zhongqi Li, Qi Zhang, Meng Xiao, Bo Shen, Lin Li, Hao Yu, Li Yan, Pingyi Zhou, Xin Wang, Yuchi Ma, Ignacio Iacobacci, Yasheng Wang, Guangtai Liang, Jiansheng Wei, Xin Jiang, Qianxiang Wang, Qun Liu

We present PanGu-Coder, a pretrained decoder-only language model adopting the PanGu-Alpha architecture for text-to-code generation, i.e. the synthesis of programming language solutions given a natural language problem description. We train PanGu-Coder using a two-stage strategy: the first stage employs Causal Language Modelling (CLM) to pre-train on raw programming language data, while the second stage uses a combination of Causal Language Modelling and Masked Language Modelling (MLM) training objectives that focus on the downstream task of text-to-code generation and train on loosely curated pairs of natural language program definitions and code functions. Finally, we discuss PanGu-Coder-FT, which is fine-tuned on a combination of competitive programming problems and code with continuous integration tests. We evaluate PanGu-Coder with a focus on whether it generates functionally correct programs and demonstrate that it achieves equivalent or better performance than similarly sized models, such as CodeX, while attending a smaller context window and training on less data.

📄 PDF Abstract BibTeX arXiv:2207.11280

Code (1)

MindSpore-paper-code-2/code399/tree/main/Pangu_alpha mindspore

Tasks

Code GenerationDecoderLanguage ModelingLanguage ModellingProgram SynthesisText-to-Code Generation

Similar Papers 제목 키워드 기반

PanGu-Draw: Advancing Resource-Efficient Text-to-Image Synthesis with Time-Decoupled Training and Reusable Coop-Diffusion

2023-12-27 · Guansong Lu, Yuanfan Guo, Jianhua Han, Minzhe Niu 외

Current large-scale diffusion models represent a giant leap forward in conditional image synthesis, capable of interpreting diverse cues like text, human poses, and edges. However, their reliance on substantial computati…

Computational EfficiencyDenoisingImage Generation

PanGu-Coder2: Boosting Large Language Models for Code with Ranking Feedback

2023-07-27 · Bo Shen, Jiaxin Zhang, Taihong Chen, Daoguang Zan 외

Large Language Models for Code (Code LLM) are flourishing. New and powerful models are released on a weekly basis, demonstrating remarkable performance on the code generation task. Various approaches have been proposed t…

Code GenerationHumanEval

Autoencoders as Tools for Program Synthesis

2021-08-16 · Sander de Bruin, Vadim Liventsev, Milan Petković

Recently there have been many advances in research on language modeling of source code. Applications range from code suggestion and completion to code summarization. However, complete program synthesis of industry-grade …

Code SummarizationLanguage ModelingLanguage ModellingProgram Synthesis

Neural Program Synthesis with a Differentiable Fixer

2020-06-19 · Matej Balog, Rishabh Singh, Petros Maniatis, Charles Sutton

We present a new program synthesis approach that combines an encoder-decoder based synthesis architecture with a differentiable program fixer. Our approach is inspired from the fact that human developers seldom get their…

DecoderProgram Synthesis

$\mathcal{B}$-Coder: Value-Based Deep Reinforcement Learning for Program Synthesis

2023-10-04 · Zishun Yu, Yunzhe Tao, Liyu Chen, Tao Sun 외

Program synthesis aims to create accurate, executable programs from problem specifications, specifically from natural language descriptions in our context. Recent studies have leveraged the power of reinforcement learnin…

Code GenerationDeep Reinforcement LearningProgram Synthesisreinforcement-learning+1