Yunrui Pei
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Yunrui Pei has a total of 15 co-authors. Here are the top co-authors by number of shared publications:
B.-M. Feng
C School of Life Sciences and Biotechnology, Dalian University , Dalian , China
1 shared publication- Bo Jiang
ByteDance, Shenzhen, China
1 shared publication - Chenyan Liu
National University of Singapore, Singapore, Singapore
1 shared publication G. Chen
A Key Laboratory of Structure-Based Drug Design & Discovery Ministry of Education , Shenyang Pharmaceutical University , Shenyang , China
1 shared publicationH.-M. Hua
A Key Laboratory of Structure-Based Drug Design & Discovery Ministry of Education , Shenyang Pharmaceutical University , Shenyang , China
1 shared publication- Haifeng Wang
Department of Traditional Chinese Materia Medica, Shenyang Pharmaceutical University, Shenyang 110016, China
1 shared publication
Latest Publications
- Open Access
- 11 September 2024
- Published by Association for Computing Machinery (ACM)
- p. 466-478
CoEdPilot: Recommending Code Edits with Learned Prior Edit Relevance, Project-wise Awareness, and Interactive Nature
Recent years have seen the development of LLM-based code generation. Compared to generating code in a software project, incremental code edits are empirically observed to be more frequent. The emerging code editing approaches usually formulate the problem as generating an edit based on known relevant prior edits and context. However, practical code edits can be more complicated. First, an editing session can include multiple (ir)relevant edits to the code under edit. Second, the inference of the subsequent edits is non-trivial as the scope of its ripple effect can be the whole project. In this work, we propose CoEdPilot, an LLM-driven solution to recommend code edits by discriminating the relevant edits, exploring their interactive natures, and estimating its ripple effect in the project. Specifically, CoEdPilot orchestrates multiple neural transformers to identify what and how to edit in the project regarding both edit location and edit content. When a user accomplishes an edit with an optional editing description, an Subsequent Edit Analysis first reports the most relevant files in the project with what types of edits (e.g., keep, insert, and replace) can happen for each line of their code. Next, an Edit-content Generator generates concrete edit options for the lines of code, regarding its relevant prior changes reported by an Edit-dependency Analyzer. Last, both the Subsequent Edit Analysis and the Edit-content Generator capture relevant prior edits as feedback to readjust their recommendations. We train our models by collecting over 180K commits from 471 open-source projects in 5 programming languages. Our extensive experiments show that (1) CoEdPilot can well predict the edits (i.e., predicting edit location with accuracy of 70.8%-85.3%, and the edit content with exact match rate of 41.8% and BLEU4 score of 60.7); (2) CoEdPilot can well boost existing edit generators such as GRACE and CCT5 on exact match rate by 8.57% points and BLEU4 score by 18.08. Last, our user study on 18 participants with 3 editing tasks (1) shows that CoEdPilot can be effective in assisting users to edit code in comparison with Copilot, and (2) sheds light on the future improvement of the tool design. The video demonstration of our tool is available at https://sites.google.com/view/coedpilot/home.
- 11 September 2024
- Published by Association for Computing Machinery (ACM)
- p. 466-478
Open AccessCoEdPilot: Recommending Code Edits with Learned Prior Edit Relevance, Project-wise Awareness, and Interactive Nature
Recent years have seen the development of LLM-based code generation. Compared to generating code in a software project, incremental code edits are empirically observed to be more frequent. The emerging code editing approaches usually formulate the problem as generating an edit based on known relevant prior edits and context. However, practical code edits can be more complicated. First, an editing session can include multiple (ir)relevant edits to the code under edit. Second, the inference of the subsequent edits is non-trivial as the scope of its ripple effect can be the whole project. In this work, we propose CoEdPilot, an LLM-driven solution to recommend code edits by discriminating the relevant edits, exploring their interactive natures, and estimating its ripple effect in the project. Specifically, CoEdPilot orchestrates multiple neural transformers to identify what and how to edit in the project regarding both edit location and edit content. When a user accomplishes an edit with an optional editing description, an Subsequent Edit Analysis first reports the most relevant files in the project with what types of edits (e.g., keep, insert, and replace) can happen for each line of their code. Next, an Edit-content Generator generates concrete edit options for the lines of code, regarding its relevant prior changes reported by an Edit-dependency Analyzer. Last, both the Subsequent Edit Analysis and the Edit-content Generator capture relevant prior edits as feedback to readjust their recommendations. We train our models by collecting over 180K commits from 471 open-source projects in 5 programming languages. Our extensive experiments show that (1) CoEdPilot can well predict the edits (i.e., predicting edit location with accuracy of 70.8%-85.3%, and the edit content with exact match rate of 41.8% and BLEU4 score of 60.7); (2) CoEdPilot can well boost existing edit generators such as GRACE and CCT5 on exact match rate by 8.57% points and BLEU4 score by 18.08. Last, our user study on 18 participants with 3 editing tasks (1) shows that CoEdPilot can be effective in assisting users to edit code in comparison with Copilot, and (2) sheds light on the future improvement of the tool design. The video demonstration of our tool is available at https://sites.google.com/view/coedpilot/home.
- 8 September 2016
- Published by Taylor & Francis in Xenobiotica
- Vol. 47 (3) , 236-244
Isolation and identification of phase I metabolites of butyrolactone I in rats
1. Butyrolactone I (BL-I), one of the major secondary metabolites of fungus Aspergillus terreus, is a selective cdc2 kinase inhibitor. In the present study, the metabolism of BL-I in male Wistar rats was investigated by characterizing metabolites excreted into feces. 2. Following an oral dose of 40 mg/kg BL-I, 10 phase I metabolites were isolated from the feces of rats, and their structures were identified on the basis of a range of spectroscopic data and ICD analysis. These metabolites were fully characterized as butyrolactone VI (M1), aspernolide E (M2), 7′′S-hydroxy-9′′-ene-butyrolactone I (M3), 7′′R-hydroxy-9′′-ene-butyrolactone I (M4), 7″S, 8″R-dihydroxy-aspernolide E (M5), 7″R, 8″S-dihydroxy-aspernolide E (M6), 7″R-acetyl-8″S-hydroxy-aspernolide E (M7), 7″S-acetyl-8″R-hydroxy-aspernolide E (M8), 7″R-methoxy-8″S-hydroxy-aspernolide E (M9), butyrolactone V (M10), respectively. It is the first time to describe the metabolites of BL-I in vivo, and metabolites M3 to M9 are new compounds. 3. BL-I and metabolites M2 to M10 were evaluated for their antimicrobial activity and in vitro antiproliferative activities. Only M-3 and M-4 showed inhibitory effect against staphylococcus aureus both with MIC of 125 μg/ml. BL-I and metabolites M-4 and M-5 exhibited potent cancer cell growth inhibitory activities against HL-60 (human leukemia) cell lines with the IC50 values of 13.2, 28.8 and 35.7 μM, respectively. 4. On the basis of metabolites profile, a possible metabolism pathway for BL-I in rats has been proposed. This is the first systematic study on the phase I metabolites of BL-I.
- 8 September 2016
- Published by Taylor & Francis in Xenobiotica
- Vol. 47 (3) , 236-244
Isolation and identification of phase I metabolites of butyrolactone I in rats
1. Butyrolactone I (BL-I), one of the major secondary metabolites of fungus Aspergillus terreus, is a selective cdc2 kinase inhibitor. In the present study, the metabolism of BL-I in male Wistar rats was investigated by characterizing metabolites excreted into feces. 2. Following an oral dose of 40 mg/kg BL-I, 10 phase I metabolites were isolated from the feces of rats, and their structures were identified on the basis of a range of spectroscopic data and ICD analysis. These metabolites were fully characterized as butyrolactone VI (M1), aspernolide E (M2), 7′′S-hydroxy-9′′-ene-butyrolactone I (M3), 7′′R-hydroxy-9′′-ene-butyrolactone I (M4), 7″S, 8″R-dihydroxy-aspernolide E (M5), 7″R, 8″S-dihydroxy-aspernolide E (M6), 7″R-acetyl-8″S-hydroxy-aspernolide E (M7), 7″S-acetyl-8″R-hydroxy-aspernolide E (M8), 7″R-methoxy-8″S-hydroxy-aspernolide E (M9), butyrolactone V (M10), respectively. It is the first time to describe the metabolites of BL-I in vivo, and metabolites M3 to M9 are new compounds. 3. BL-I and metabolites M2 to M10 were evaluated for their antimicrobial activity and in vitro antiproliferative activities. Only M-3 and M-4 showed inhibitory effect against staphylococcus aureus both with MIC of 125 μg/ml. BL-I and metabolites M-4 and M-5 exhibited potent cancer cell growth inhibitory activities against HL-60 (human leukemia) cell lines with the IC50 values of 13.2, 28.8 and 35.7 μM, respectively. 4. On the basis of metabolites profile, a possible metabolism pathway for BL-I in rats has been proposed. This is the first systematic study on the phase I metabolites of BL-I.