Výsledky vyhledávání

Report

FastFixer: An Efficient and Effective Approach for Repairing Programming Assignments

Autor: Liu, Fang, Liu, Zhenwei, Zhao, Qianhui, Jiang, Jing, Zhang, Li, Li, Ge, Sun, Zian, Li, Zhongqi, Ma, Yuchi

Providing personalized and timely feedback for student's programming assignments is useful for programming education. Automated program repair (APR) techniques have been used to fix the bugs in programming assignments, where the Large Language Models

Externí odkaz: http://arxiv.org/abs/2410.21285

Zobrazit plný text záznamu

Report

Iterative Knowledge Distillation through Feedback-Driven Learning Cycles

Autor: Chen, Yujia, Ye, Yang, Li, Zhongqi, Ma, Yuchi, Gao, Cuiyun

Large code models (LCMs) have remarkably advanced the field of code intelligence. Despite their impressive capabilities, they still face practical employment challenges, such as high costs, limited accessibility of proprietary LCMs, and adaptability

Externí odkaz: http://arxiv.org/abs/2408.03680

Zobrazit plný text záznamu

Report

Exploring and Evaluating Hallucinations in LLM-Powered Code Generation

Autor: Liu, Fang, Liu, Yang, Shi, Lin, Huang, Houkun, Wang, Ruifeng, Yang, Zhen, Zhang, Li, Li, Zhongqi, Ma, Yuchi

The rise of Large Language Models (LLMs) has significantly advanced many applications on software engineering tasks, particularly in code generation. Despite the promising performance, LLMs are prone to generate hallucinations, which means LLMs might

Externí odkaz: http://arxiv.org/abs/2404.00971

Zobrazit plný text záznamu

Report

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

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-Code

Externí odkaz: http://arxiv.org/abs/2207.11280

Zobrazit plný text záznamu