Guoxin Yu

About me

I am currently an Assistant Researcher at Peng Cheng Laboratory. I received my Ph.D. from the Key Lab of Intelligent Information Processing, Institute of Computing Technology (ICT), Chinese Academy of Sciences (CAS) , where I was advised by Prof. Xiang Ao.
My research focuses on retrieval-augmented text classification and its applications in financial sentiment analysis. This is my Google Scholar.
I received my B.S. degree in Computer Science from Shandong Normal University in 2020.
From April 2022 to May 2023, I interned at the NLP Center, Tencent AI Lab, under the supervision of Dr. Lemao Liu.

Research

Retrieval-augmented natural language understanding.
Retrieval-augmented natural language understanding aims to fetch relevant examples from training sets or external knowledge bases to improve the performance of different NLU tasks. The main challenge of this task is to retrieve useful examples for downstream tasks.
Aspect-based Sentiment Analysis.
Aspect-based Sentiment Analysis (ABSA for short) is a burgeoning fine-grained sentiment analysis task, which aims to extract aspect terms, classify the related sentiment polarities, and find opinion terms as the cause of the sentiment.
Financial Event Extraction and Classification.
Financial Event extraction and Classification extract events form financial news and reports and obtain their classes. Such task could help users obtain competitors’ strategies, predict the stock market and make correct investment decisions.

Publications

Rethinking the Evaluation of In-Context Learning for LLMs. Accepted by EMNLP2024.
Guoxin Yu, Lemao Liu, Mo Yu, Yue Yu, and Xiang Ao.
Improving Event-Level Financial Sentiment Analysis with Retrieval-Augmented Multipath Chain-of-Thought Prompting. CCF BigData 2024.
Yiming Zhang, Xiang Ao, Guoxin Yu, and Qing He.
EFSA: Towards Event-Level Financial Sentiment Analysis. In the Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (ACL2023). [Paper]
Tianyu Chen, Yiming Zhang, Guoxin Yu, Dapeng Zhang, Li Zeng, Qing He, and Xiang Ao*.
Retrieval-augmented Few-shot Text Classification. In the 2023 Conference on Empirical Methods in Natural Language Processing (EMNLP2023 Findings). [Paper] [Slide] [Poster]
Guoxin Yu, Lemao Liu*, Haiyun Jiang, Shuming Shi, Xiang Ao*.
Making Better Use of Training Corpus: Retrieval-based Aspect Sentiment Triplet Extraction via Label Interpolation. In the Joint Conference of the 61th Annual Meeting of the Association for Computational Linguistics and the 13th International Joint Conference on Natural Language Processing (ACL-IJCNLP2023 Findings). [Paper] [Slide] [Poster]
Guoxin Yu, Lemao Liu*, Haiyun Jiang, Shuming Shi, and Xiang Ao*.
Self Question-answering: Aspect-based Sentiment Analysis by Role Flipped Machine Reading Comprehension. In the 2021 Conference on Empirical Methods in Natural Language Processing (EMNLP2021 Findings). [Paper] [Slide]
Guoxin Yu, Jiwei Li, Ling Luo, Yuxian Meng, Xiang Ao* and Qing He.
Making Flexible Use of Subtasks: A Multiplex Interaction Network for Unified Aspect-based Sentiment Analysis. In the Joint Conference of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (ACL-IJCNLP2021 Findings). [Paper] [Slide]
Guoxin Yu, Xiang Ao*, Ling Luo, Min Yang, Xiaofei Sun, Jiwei Li and Qing He.
Discovering Protagonist of Sentiment with Aspect Reconstructed Capsule Network. In Proceedings of the 26th International Conference on Database Systems for Advanced Applications. (DASFAA2021).
[Paper] [Slide]
Chi Xu, Hao Feng, Guoxin Yu, Min Yang, Xiting Wang, Yan Song and Xiang Ao*.

Guoxin Yu 于国欣

About me

Research

Publications

Honours and Awards