Web IR NLP Group @ NUS
Web IR NLP Group @ NUS
Tour
News
People
Events
Publications
Projects
Contact
Light
Dark
Automatic
NLP
Navigating Ethical Challenges in NLP: Hands-on strategies for students and researchers (Tutorial at ACL '25)
Navigating Ethical Challenges in NLP
27 Jul, 2025
Austria Center Vienna
Min-Yen Kan
,
Minzhi Li (Ella)
A Comparative Analysis of Representation Flow in State-Space and Transformer Architectures
We study the feature flow across layers in state-space models compared to Transformers.
Nhat M. Hoang
,
Xuan Long Do
,
Anh Tuan Luu
,
Min-Yen Kan
ChatCRS: Incorporating External Knowledge and Goal Guidance for LLM-based Conversational Recommender Systems
We incoporate external knowledge and goals for better CRS.
Chuang Li
,
Hengchang Hu
,
Min-Yen Kan
PDF
Code
Summarizing Scientific Content with Social Media Language
Translating complex scientific content into accessible language for social media platforms.
Amy Ling
,
Jiaying Wu
,
Min-Yen Kan
Assessing the AI Scientist
We find positive use cases, but also weaknesses that need much handholding from human scientists.
Joeran Beel
,
Moritz Baumgart
,
Min-Yen Kan
,
Haomeng Lu
PDF
Discourse and LLMs
We measure and enhance LLMs’ discourse understanding and reasoning faithfulness.
Yisong Miao
,
Min-Yen Kan
PDF
Code
Slides
Benchmarking and Improving OCR Systems for Southeast Asian Languages
We benchmark and improve modern OCR systems on Southeast Asian languages.
Jason Qiu
,
Tongyao Zhu
,
Min-Yen Kan
PDF
Code
Slides
Linguistics Informed Natural Language Processing with Graph Networks
Question-Answering with Linguistics-Based Graph Attention Networks.
Shirshajit Sen Gupta
,
Min-Yen Kan
Code
Slides
Emoji Composition
We provide language resource for emoji composition.
Yisong Miao
,
Zi Yun Yang
,
Min-Yen Kan
PDF
Code
Slides
Perspectives on crowdsourcing annotations for natural language processing
Crowdsourcing has emerged as a new method for obtaining annotations for training models for machine learning. While many variants of …
Aobo Wang
,
Cong Duy Vu Hoang
,
Min-Yen Kan
Cite
DOI
URL
»
Cite
×