Wei Li

School of Computer Science

Peking University

Hi! I am a PhD student of computer science at the Peking University. My research interests include repository-level code generation and grammatical error correction.

In the admission year of 2022, I entered the Computational Linguistics Lab in the Peking University as a Master Student. And my research field was Grammatical Error Correction. I have 2 papers (ACL 2024; NAACL 2025) in this field.

In the 2024, I applied for changing the target from Master’s degree to the PhD degree. And my main research field changes to Code Generation under repository scenario. I have published FEA-Bench (ACL 2025) and for now I am conducting research on post-training of LLMs for repository-level code development.

Feel free to contact me if you are interested in the code generation of LLMs. Welcome any collabrations.

I am interested in linguistics and languages. For now, except English and Chinese, my Japnanese (N2) and Korean (TOPIK-5) is above medium-level. I also have experience in learning Français, Deutsch, Portugues, Italian, Arab, Russian languages.

In my daily life, sports and photography are also my hobbies. I mainly play tennis, and is around 3.0 level. If you live around Peking University, welcome to make an appointment for tennis, squash, badminton, and so on.

news

Jul 25, 2025	We launched the official website of the FEA-Bench! https://gmago-leway.github.io/fea-bench.github.io/
May 22, 2025	✨4 Papers are accepted by ACL 2024. (3 Main Conference, 1 Findings)

selected publications

FEA-Bench: A Benchmark for Evaluating Repository-Level Code Generation for Feature Implementation

Wei Li, Xin Zhang, Zhongxin Guo, and 6 more authors

In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), Jul 2025
Explanation based In-Context Demonstrations Retrieval for Multilingual Grammatical Error Correction

Wei Li, Wen Luo, Guangyue Peng, and 1 more author

In Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers), Apr 2025

DOI
Detection-Correction Structure via General Language Model for Grammatical Error Correction

Wei Li and Houfeng Wang

In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), Aug 2024

DOI
Transformer-based Feature Reconstruction Network for Robust Multimodal Sentiment Analysis

Ziqi* Yuan, Wei* Li, Hua Xu, and 1 more author

In Proceedings of the 29th ACM International Conference on Multimedia, Virtual Event, China, Aug 2021

Abs DOI

Improving robustness against data missing has become one of the core challenges in Multimodal Sentiment Analysis (MSA), which aims to judge speaker sentiments from the language, visual, and acoustic signals. In the current research, translation-based methods and tensor regularization methods are proposed for MSA with incomplete modality features. However, both of them fail to cope with random modality feature missing in non-aligned sequences. In this paper, a transformer-based feature reconstruction network (TFR-Net) is proposed to improve the robustness of models for the random missing in non-aligned modality sequences. First, intra-modal and inter-modal attention-based extractors are adopted to learn robust representations for each element in modality sequences. Then, a reconstruction module is proposed to generate the missing modality features. With the supervision of SmoothL1Loss between generated and complete sequences, TFR-Net is expected to learn semantic-level features corresponding to missing features. Extensive experiments on two public benchmark datasets show that our model achieves good results against data missing across various missing modality combinations and various missing degrees.