PEACE: Towards Efficient Project-Level Performance Optimization via Hybrid Code Editing
This program is tentative and subject to change.
Large Language Models (LLMs) have demonstrated significant capability in code generation, but their potential in code optimization remains underexplored. Previous LLM-based code optimization approaches exclusively focus on function-level optimization and overlook interaction between functions, failing to generalize to real-world development scenarios. Code editing techniques show great potential for conducting project-level code optimization, yet they face challenges associated with invalid edits and suboptimal internal functions. To address these gaps, we propose PEACE, a novel hybrid framework for \textbf{P}roject-level p\textbf{E}rformance optimization through \textbf{A}utomatic \textbf{C}ode \textbf{E}diting, which also ensures the overall correctness and integrity of the project. PEACE integrates three key phases: dependency-aware optimizing function sequence construction, valid associated edits identification, and performance editing iteration. To rigorously evaluate the effectiveness of PEACE, we construct PEACExec, the first benchmark comprising 146 real-world optimization tasks from 47 high-impact GitHub Python projects, along with highly qualified test cases and executable environments. Extensive experiments demonstrate PEACE’s superiority over the state-of-the-art baselines, achieving a 69.2% correctness rate (pass@1) and +46.9% opt rate in execution efficiency. Notably, our PEACE outperforms all baselines by significant margins, particularly in complex optimization tasks with multiple functions. Moreover, extensive experiments are also conducted to validate the contributions of each component in PEACE, as well as the rationale and effectiveness of our hybrid framework design.
This program is tentative and subject to change.
Mon 17 NovDisplayed time zone: Seoul change
11:00 - 12:30 | |||
11:00 12mTalk | AutoFid: Adaptive and Noise-Aware Fidelity Measurement for Quantum Programs via Circuit Graph Analysis Research Papers | ||
11:12 12mTalk | HybridSIMD: A Super C++ SIMD Library with Integrated Auto-tuning Capabilities Research Papers Haolin Pan Institute of Software, Chinese Academy of Sciences;School of Intelligent Science and Technology, HIAS, UCAS, Hangzhou;University of Chinese Academy of Sciences, Xulin Zhou Institute of Software, Chinese Academy of Sciences; University of Chinese Academy of Sciences, Mingjie Xing Institute of Software, Chinese Academy of Sciences, Yanjun Wu Institute of Software, Chinese Academy of Sciences | ||
11:25 12mTalk | PEACE: Towards Efficient Project-Level Performance Optimization via Hybrid Code Editing Research Papers Xiaoxue Ren Zhejiang University, Jun Wan Zhejiang University, Yun Peng The Chinese University of Hong Kong, Zhongxin Liu Zhejiang University, Ming Liang Ant Group, Dajun Chen Ant Group, Wei Jiang Ant Group, Yong Li Ant Group | ||
11:38 12mTalk | CoTune: Co-evolutionary Configuration Tuning Research Papers Gangda Xiong University of Electronic Science and Technology of China, Tao Chen University of Birmingham Pre-print | ||
11:51 12mTalk | It's Not Easy Being Green: On the Energy Efficiency of Programming Languages Research Papers Nicolas van Kempen University of Massachusetts Amherst, USA, Hyuk-Je Kwon University of Massachusetts Amherst, Dung Nguyen University of Massachusetts Amherst, Emery D. Berger University of Massachusetts Amherst and Amazon Web Services | ||
12:04 12mTalk | When Faster Isn't Greener: The Hidden Costs of LLM-Based Code Optimization Research Papers Tristan Coignion Université de Lille - Inria, Clément Quinton Université de Lille, Romain Rouvoy University Lille 1 and INRIA | ||
12:17 12mTalk | United We Stand: Towards End-to-End Log-based Fault Diagnosis via Interactive Multi-Task Learning Research Papers Minghua He Peking University, Chiming Duan Peking University, Pei Xiao Peking University, Tong Jia Institute for Artificial Intelligence, Peking University, Beijing, China, Siyu Yu The Chinese University of Hong Kong, Shenzhen (CUHK-Shenzhen), Lingzhe Zhang Peking University, China, Weijie Hong Peking university, Jing Han ZTE Corporation, Yifan Wu Peking University, Ying Li School of Software and Microelectronics, Peking University, Beijing, China, Gang Huang Peking University | ||