What doesn’t kill you makes you stronger.

I am Runxin Xu (许润昕), a master student at the Institute of Computational Linguistics in the School of EECS, Peking University, advised by Dr. Baobao Chang and Dr. Zhifang Sui. Prior to this, I earned my Bachelor’s degree at Shanghai Jiao Tong University advised by Dr. Xiaofeng Gao.

My research interests mainly lie in natural language processing, especially 1) document-level and few-shot information extraction, and 2) effective and efficient pre-trained language model.

I am actively seeking for engineer or researcher position (graduate in 2023). Please drop me an email (A@B,A=runxinxu,B=gmail.com) if you want to chat.


Peking University, Sep. 2020 - Jun. 2023 (expected)

Shanghai Jiao Tong University, Sep. 2016 - Jun. 2020


ByteDance Search, *December. 2021 – Now *

  • E-Commerce in Douyin Search.
  • Advised by Shian Chen, Zhe Chen, and Pengcheng Yang.

Alibaba Damo Academy, Mar. 2021 – December. 2021

ByteDance AI Lab, Nov. 2019 – Jan. 2021

Microsoft C+AI, July. 2019 – October. 2019

Publications and Manuscripts

* means equal contribution.

From Dense to Sparse: Contrastive Pruning for Better Pre-trained Language Model Compression. [PDF] [code]
Runxin Xu*, Fuli Luo*, Chengyu Wang, Baobao Chang, Jun Huang, Songfang Huang, Fei Huang.
AAAI2022 (Oral).

S4-Tuning: A Simple Cross-lingual Sub-network Tuning Method.
Runxin Xu, Fuli Luo, Baobao Chang, Songfang Huang, Fei Huang.

A Two-Stream AMR-enhanced Model for Document-level Event Argument Extraction. [PDF] [code]
Runxin Xu*, Peiyi Wang*, Tianyu Liu, Shuang Zeng, Baobao Chang, Zhifang Sui.
NAACL2022 (Oral).

An Enhanced Span-based Decomposition Method for Few-Shot Sequence Labeling. [PDF] [code]
Peiyi Wang*, Runxin Xu*, Tianyu Liu, Qingyu Zhou, Yunbo Cao, Baobao Chang, Zhifang Sui.
NAACL2022 (Oral).

BERT Raises a Child: Towards Improving Generalization for Large Language Model Fine-tuning. [PDF] [code] [Report]
Runxin Xu*, Fuli Luo*, Zhiyuan Zhang, Chuanqi Tan, Baobao Chang, Songfang Huang, Fei Huang.
EMNLP2021 (Oral).

Behind the Scenes: An Exploration of Trigger Biases Problem in Few-Shot Event Classification. [PDF] [code]
Peiyi Wang*, Runxin Xu*, Tianyu Liu, Damai Dai, Baobao Chang, Zhifang Sui.

Document-level Event Extraction via Heterogeneous Graph-based Interaction Model with a Tracker. [PDF] [code] [talk]
Runxin Xu, Tianyu Liu, Lei Li, Baobao Chang.
ACL2021 (Oral).

Double Graph Based Reasoning for Document-level Relation Extraction. [PDF] [code]
Shuang Zeng*, Runxin Xu*, Baobao Chang, Lei Li.

Volctrans Parallel Corpus Filtering System for WMT 2020. [PDF]
Runxin Xu, Zhuo Zhi, Jun Cao, Mingxuan Wang, Lei Li.
(Win the first place of WMT20 Parallel Corpus Filtering tasks. Here is the result.)

Xiaomingbot: A Multilingual Robot News Reporter [PDF] [Report] [URL] [Youtube]
Runxin Xu, Jun Cao, Mingxuan Wang, Jiaze Chen, Hao Zhou, Ying Zeng, Yuping Wang, Li Chen, Xiang Yin, Xijin Zhang, Songcheng Jiang, Yuxuan Wang, Lei Li.
ACL2020, System Demonstrations.

Focus on the Target’s Vocabulary: Masked Label Smoothing for Machine Translation. [PDF] [code]
Liang Chen, Runxin Xu, Baobao Chang.

Probing Structured Pruning on Multilingual Pre-trained Models: Setting, Algorithm, and Efficiency. [PDF]
Yanyang Li, Fuli Luo, Runxin Xu, Songfang Huang, Fei Huang, Liwei Wang.

A Double-Graph Based Framework for Frame Semantic Parsing. [code]
Ce Zheng, Xudong Chen, Runxin Xu, Baobao Chang.

ATP: AMRize Then Parse! Enhancing AMR Parsing with PseudoAMRs. [PDF] [code]
Liang Chen, Peiyi Wang, Runxin Xu, Tianyu Liu, Zhifang Sui, Baobao Chang.

ACMo: Angle-Calibrated Moment Methods for Stochastic Optimization. [PDF] [code] [URL]
Xunpeng Huang, Runxin Xu, Hao Zhou, Zhe Wang, Zhengyang Liu, Lei Li.

Making Pre-trained Language Models End-to-end Few-shot Learners with Contrastive Prompt Tuning. [PDF]
Ziyun Xu*, Chengyu Wang*, Minghui Qiu, Fuli Luo, Runxin Xu, Songfang Huang, Jun Huang

Explicit Interaction Network for Aspect Sentiment Triplet Extraction. [PDF]
Peiyi Wang*, Lianzhe Huang*, Tianyu Liu, Damai Dai, Runxin Xu, Houfeng Wang, Baobao Chang, Zhifang Sui.

Adaptive Gradient Methods Can Be Provably Faster than SGD after Finite Epochs. [PDF]
Xunpeng Huang, Hao Zhou, Runxin Xu, Zhe Wang, Lei Li.

Hornors & Awards

Teaching Experience

  • Introduction to Computing (Category C), Teaching Assistant, Fall 2020

Contact Me

  • Email: runxinxu AT gmail DOT com
  • Links: GitHub