|
I am a Ph.D. student of Chinese Information Processing Center, Institute of Software, Chinese Academy of Sciences, supervised by Le Sun.
In September 2022, I was admitted to study for a M.Sc. degree without entrance examination.
In July 2023, I successfully received my Bachelor of Engineering degree from School of Software Engineering of South China University of Technology.
Contact: zhengjiasheng2022{at}iscas{dot}ac{dot}cn |
|
|
| 04/2026 | ScaleBox is accepted by ACL 2026 Demo |
| 04/2026 | Officially recognized as a CANN (Huawei Ascend) Core Contributor |
| 12/2025 | The first official long-context code RL practice merged into CANN main repository. |
| 07/2024 | Release Beyond Correctness: Benchmarking Multi-dimensional Code Generation for Large Language Models |
|
|
|
|
|
Jiasheng Zheng, Xin Zheng, Boxi Cao, Pengbo Wang, Zhengzhao Ma, Qiming Zhu, Jiazhen Jiang, Yaojie Lu, Hongyu Lin, Xianpei Han, Le Sun ACL 2026 Demo Github |
|
|
Jiasheng Zheng, Boxi Cao, Zhengzhao Ma, Ruotong Pan, Hongyu Lin, Yaojie Lu, Xianpei Han, Le Sun arXiv preprint arXiv: 2407.11470 (2024) ArXiv / Github / Leaderboard |
|
Jiasheng Zheng, Hongyu Lin, Boxi Cao, Meng Liao, Yaojie Lu, Xianpei Han, Le Sun arXiv preprint arXiv: 2410.07693 (2024) ArXiv |
|
Jiayuan Xie, Jiasheng Zheng, Wenhao Fang, Yi Cai, Qing Li Neural Networks, 184 (2025) Paper |
|
|
|
|
|
First Official Long-Context Code RL Training Infrastructure on Huawei Ascend
TL;DR: Based on verl and ScaleBox, we validated the effectiveness of Code RL on 1.5B, 4B, and 30B-A3B model scales. This contribution was recognized by the Ascend team and merged into the main repository. Gitcode / WeChat Live |
|
ScaleBox: A Scalable Sandbox for Distributed Code Execution, RL Training and Unified Benchmarking
TL;DR: It supports both NVIDIA and Ascend platforms, and improves the fidelity and efficiency of code validation. Github / WeChat Article |
|
ReasoningLens: A User-friendly Toolkit to Visualize, Understand, and Debug Model Reasoning Chains
TL;DR: Out-of-the-box and integrated into Open WebUI. Github / Blog |
|
|
| Beike |
LLM Algorithm Engineer, Beijing, China, 10/2024 - 06/2025
Alignment & Knowledge Injection in LLMs for the home decoration domain |
|
|
| 2026 |
CANN (Huawei Ascend) Core Contributor
|
| 2024 | University of Chinese Academy of Sciences (UCAS) Outstanding Student |
| 2020, 2021, 2022 | South China University of Technology (SCUT) Second-Class Scholarship |
| 2020 | South China University of Technology (SCUT) Outstanding Student |