Web IR / NLP Group at the National University of Singapore

Singapore
Joined July 2012
Our paper "AnalysisBank: An Expert Analysis Pattern Library for Financial Report Generation" will be presented as an Oral Presentation at #EMNLP2026! 🚀📊 💡 TL;DR: We introduce a structured expert analysis pattern library to significantly improve how LLMs generate financial reports with data-specific analysis and in-depth insight. 📅 Session Details: October 27, 2026 | 11:00 – 12:30 | Room F8 📄 Paper: arxiv.org/abs/2609.00818 💻 Code & Dataset: github.com/yajingyang/Analys… Work by @y_yajing, Kelvin J. L. Koa, and @knmnyn! Make sure to catch the talk! 🙌
1
2
141
🎓️ Congratulations to Dr Minzhi Li (Ella; @EllaMinzhiLi) on successfully defending her PhD thesis, 🌟Towards Language Technology Evaluation Grounded in Human Judgement🌟!
1
2
920
Many thanks to her co-advisors @NancyFChen, @ShafiqJoty and @Diyi_Yang, and to her examiners Tat-Seng Chua, Brian Lim and @BryanHooi1. Ella defended her thesis on 12 Aug 2026.
1
1
330
We look forward to seeing where her research goes next, particularly in aligning audio and multicultural models. 🔗 Learn more about Ella’s work: yocodeyo.github.io:
1
270
🌟 The Future of NLP Needs Ethical Considerations: An Interactive Tutorial for Researchers and Students🌟 📣 Please share this announcement with your colleagues!
1
2
1
4
687
(Remote AACL-IJCNLP participants are welcome to attend as well; we'll have faculty and students ready to help remote participants!)
1
36
w/ Barid Xi Ai, Vamshi Krishna Bonagiri, Abdoul Jalil Djiberou Mahamadou, @estherganz @ZhijingJin, @knmnyn, Seunghun Lee, @YisongMiao, Margot Mieskes, Aurélie Névéol, & @yajing08042
53
We were privileged to host Dr Ingo Frommholz (@ingo@idf.social) for short visit from Modul University Austria in Aug! Ingo gave a seminar entitled 🌟 "Fake Science" and Information Overload in Academia🌟 , a very real problem facing researchers needing to triage their
1
1
1
170
In this talk, we will examine the evolving problem of fake science and its role in driving information overload. We will outline recent developments in scholarly information access, highlight open research problems — from detecting low-quality and fraudulent content to designing
1
8
veracity-aware retrieval and recommendation models — and discuss how IR research can contribute to ensuring that high-quality knowledge remains discoverable, trustworthy, and actionable in an era of overwhelming information abundance.
9