Prof. Beel’s 3-months visit of Prof. Min-Yen Kan in Singapore comes to an end

As my research semester in Singapore draws to a close, I find myself reflecting on an incredibly rewarding three months at the National University of Singapore (NUS), where I had the privilege of visiting Prof. Min-Yen Kan and his outstanding Web, Information Retrieval / Natural Language Processing Group (WING). My visit has been an exciting and productive experience, marked by significant collaborations and deep dives into the rapidly evolving field of AI for Science.

Singapore is a global hub for cutting-edge research, particularly in recommender systems and information retrieval. Many top researchers are based here, including Prof. Min-Yen Kan from NUS and Prof. Aixin Sun from Nanyang Technological University (NTU). The city-state is also home to major tech companies such as Grab, Sea Group, and ByteDance, making it a dynamic ecosystem for AI and machine learning research. The presence of ACM RecSys 2023 and SIGIR 2008 in Singapore further reinforced its status as a key location for the recommender systems community.

Prof. Min-Yen Kan is a highly respected researcher in natural language processing (NLP) and information retrieval. His contributions to AI-driven academic search, citation analysis, and scientific document processing are widely recognized. His research group at NUS is equally remarkable, comprising talented PhD students who are pushing the boundaries of AI-driven knowledge discovery. One of his students recently received the prestigious Google PhD Fellowship, a testament to the group’s research excellence. I was honored to be included in this dynamic and intellectually stimulating environment. From weekly research discussions to brainstorming sessions on AI’s role in automating scientific workflows, the experience has been both inspiring and productive. Additionally, I had the opportunity to meet with many other researchers, including Aixin Sun, to discuss joint research initiatives, further strengthening academic ties between our institutions. These conversations were invaluable, as Singapore continues to foster a thriving research environment with strong collaborations between academia and industry.

A key focus of our collaboration was exploring how AI can revolutionize scientific research by automating the entire research lifecycle—from idea generation to peer review. With tools like OpenAI Deep Research and Google Deep Research demonstrating remarkable progress, we sought to critically assess and extend the state-of-the-art in AI-driven scientific discovery. One of our main research efforts examined how generative AI could autonomously generate research ideas, conduct experiments, and draft manuscripts. While existing AI tools show promise in assisting researchers, they often struggle with critical aspects such as literature retrieval, experimental validation, and producing logically sound research papers. Our work aimed to identify and mitigate these limitations, ensuring that AI-generated research meets rigorous academic standards.

We also investigated how AI can support peer review. While AI-driven review systems can provide structured feedback and consistency, our research revealed that they often exhibit biases and struggle to identify deep methodological flaws. This raises important questions about integrating AI into academic publishing workflows while maintaining human oversight. Beyond manuscript generation and review, we explored AI’s potential in reproducibility studies. AI-driven tools could be leveraged to validate research findings, ensuring that published results are replicable. However, for this to be effective, AI systems need to improve their ability to interpret and execute complex experimental setups reliably.

Thank you to everyone who made this experience so rewarding—I look forward to returning soon!