SelfPiCo: Self-Guided Partial Code Execution with LLMs
This program is tentative and subject to change.
Code executability plays a vital role in software debugging and testing (e.g., detecting runtime exceptions or assertion violations). However, code execution, especially partial or arbitrary code execution, is a non-trivial task due to missing definitions and complex third-party dependencies. To make partial code (such as code snippets posted on the web or code fragments deep inside complex software projects) executable, the existing study has proposed a machine learning model to predict the undefined element types and inject the pre-defined dummy values into execution. However, the performance of their tool is limited due to its simply designed dummy values and the inability to continue learning. In this paper, we design and implement a novel framework, named SelfPiCo (Self-Guided Partial Code Executor), to dynamically guide partial code execution by incorporating the open-source LLM (i.e., Code Llma) within an interactive loop. Particularly, SelfPiCo leverages few-shot in-context learning and chain-of-thought reasoning to elicit human knowledge and logical reasoning based on fine-tuning the Code Llama model. SelfPiCo continuously learns from code execution results and refines its predictions step after step. Our evaluations demonstrate that SelfPiCo can execute 57.3% and 71.1% of all lines in the open-source code and Stack Overflow snippets, outperforming the most recent state-of-the-art Lexecutor by 13.2% and 16.6%, respectively. Moreover, SelfPiCo successfully detected 18 and 33 runtime type error issues by executing the partial code from eight GitHub software projects and 43 Stack Overflow posts, demonstrating the practical usage and potential application of our framework in practice.
This program is tentative and subject to change.
Thu 19 SepDisplayed time zone: Amsterdam, Berlin, Bern, Rome, Stockholm, Vienna change
13:30 - 14:50 | |||
13:30 20mTalk | MicroRes: Versatile Resilience Profiling in Microservices via Degradation Dissemination Indexing Technical Papers Tianyi Yang The Chinese University of Hong Kong, Cheryl Lee The Chinese University of Hong Kong, Jiacheng Shen The Chinese University of Hong Kong, Yuxin Su Sun Yat-sen University, Cong Feng Computing and Networking Innovation Lab, Huawei Cloud Computing Technology Co., Ltd, Yongqiang Yang Huawei Technologies, Michael Lyu Chinese University of Hong Kong DOI | ||
13:50 20mTalk | Feedback-Directed Partial Execution Technical Papers Ishrak Hayet North Carolina State University, Adam Scott North Carolina State University, Marcelo d'Amorim North Carolina State University | ||
14:10 20mTalk | Define-Use Guided Path Exploration for Better Forced Execution Technical Papers Dongnan He Renmin University of China, Dongchen Xie Renmin University of China, Yujie Wang Renmin University of China, Wei You Renmin University of China, Bin Liang Renmin University of China, China, Jianjun Huang Renmin University of China, China, Wenchang Shi Renmin University of China, China, Zhuo Zhang Purdue University, Xiangyu Zhang Purdue University DOI | ||
14:30 20mTalk | SelfPiCo: Self-Guided Partial Code Execution with LLMs Technical Papers Zhipeng Xue , Zhipeng Gao Shanghai Institute for Advanced Study - Zhejiang University, Shaohua Wang Central University of Finance and Economics, Xing Hu Zhejiang University, Xin Xia Huawei Technologies, Shanping Li Zhejiang University |