Why is research fund so precious for Australia academic staff? The gross Research and Development (R&D) expenditure as a percentage of GDP in Australia is just TOO low!

Our full paper, ROLeR: Effective Reward Shaping in Offline Reinforcement Learning for Recommender Systems, is accepted to ACM CIKM 2024, with Yi Zhang, Jiajun Liu and Sen Wang. Offline reward estimation is the key! Paper: arxiv.org/abs/2407.13163 Code: github.com/ArronDZhang/RO…

Today we are very happy to have Prof Chao Huang from HKU Institute of Data Science to give a seminar at Data Science group at UQ School of EECS about "Graph Foundation Model in the Era of LLMs". uq-ds-seminar.github.io/llm4graph-chao/

We invite contributions to PhD Symposium at #TheWebConf 2025. PhD students are encouraged to submit ongoing research related to broad topic of the World Wide Web. www2025.thewebconf.org/phd-symposium gabriella pasi Ruihong Qiu and Mark Sanderson, PC Co-Chairs.

Today we are very happy to have Prof Wei Jin Wei Jin from Emory University to give a seminar (uq-ds-seminar.github.io/epi-wei/) at UQ School of EECS UQ News about GNN and Epidemic Modeling and the great toolkit EpiLearn (github.com/Emory-Melody/E…)

During the busy US election time, today we have Jack Morris jxmo from Cornell University to give us an inspiring talk about Contextual Document Embeddings (uq-ds-seminar.github.io/context-jack/) at UQ School of EECS. What a brilliant idea in obtaining doc embeddings!

Yesterday, Jiahao Zhang from Penn State gave a brilliant seminar on Long-Range Meets Scalability: Unveiling a Linear-Time Graph Neural Network for Recommendation at Scale for our seminar (uq-ds-seminar.github.io/longrangerecsy…) UQ School of EECS

Today, we have the amazing Qitian Wu from Eric and Wendy Schmidt Center Massachusetts Institute of Technology (MIT) and Harvard University to give a talk UQ School of EECS about "Towards Graph Machine Learning in the Wild" (uq-ds-seminar.github.io/gnn-wild-qitia…) with some hot discussions of OOD on graph!

Yesterday, our data science seminar UQ School of EECS were hosting Feng Liu @ Melbourne from University of Melbourne to give an amazing talk on Mitigating Distribution Shifts in Using Pre-trained Vision-Language Models (uq-ds-seminar.github.io/ds-vlm-feng/)!

Today, at the Data Science Seminar UQ School of EECS, we have Tianyu Gao Tianyu Gao from Princeton NLP Group to deliver an amazing talk on "Developing Effective Long-Context Language Models" (uq-ds-seminar.github.io/ds-lclm-tianyu/)

Today, we are very happy to have Jianling Wang from Google DeepMind to give a fabulous talk on "When LLMs Meet Recommendations: Scalable Hybrid Approaches to Enhance User Experiences" (uq-ds-seminar.github.io/llmrec-jianlin…) at the data science seminar UQ School of EECS

Yesterday, we were very glad to have Yunzhi Zhang from Stanford University to give a seminar about "Controllable Visual Synthesis via Structural Representation" (uq-ds-seminar.github.io/synt-yunzhi/) at our Data Science seminar UQ School of EECS.

We invite PhD students to submit your work and join us at PhD Symposium The Web Conference 2025, (www2025.thewebconf.org/phd-symposium). The submission due date is 18 Dec, 2024 (AOE)! We will see you in the beautiful and amazing Sydney down under! #WWW2025 #WebConf2025 gabriella pasi Mark Sanderson

Our work, GOLD: Graph Out-of-Distribution Detection via Implicit Adversarial Latent Generation, is accepted to #ICLR2025 with Danny, Guangdong and Helen. Non-OOD-exposure can outperform OOD-exposure methods on graph! PDF link: openreview.net/forum?id=y5ein… Code will be released soon!

The GOLD has been accepted as a Spotlight #ICLR2025 ICLR 2026 ! The arXiv and code are both available now: arXiv: arxiv.org/abs/2502.05780 code: github.com/DannyW618/GOLD

Today, we are very happy to have Prof Heng Ji Heng Ji from UIUC NLP to give a seminar talk about Towards Knowledgeable Foundation Models UQ School of EECS ! uq-ds-seminar.github.io/knowllm-heng/

Welcome to talk to Danny about graph OOD at: Hall 3 + Hall 2B #204 Thu 24 Apr noon AEST — 2:30 p.m. AEST #ICLR2025

In our Data Science Seminar UQ School of EECS today, we are very happy to have Yupeng Hou ✈️ ICML 2025 from UC San Diego UCSD CSE to talk about the hot topic on generative RecSys and tokenization.

Today we are very happy to welcome Nicholas Tomlin from BerkeleyNLP to give a seminar talk on Reasoning with Language Models at UQ School of EECS followed by a very hot discussion!

Today we are very happy to have Sean Xuefeng Du from UW–Madison & NTU Singapore to give a seminar talk, Teach AI What It Doesn't Know, at our Data Science seminar UQ School of EECS !