New paper w/ Xiaoxin He, T. Laurent, Bryan Hooi "Explanations as Features: LLM-Based Features for Text-Attributed Graphs" ArXiv: arxiv.org/pdf/2305.19523… GitHub: github.com/XiaoxinHe/TAPE

Xiaoxin He has created a GitHub repo "Awesome-Graph-LLM" to gather research resources on the latest developments in LLM+GNN: github.com/XiaoxinHe/Awes… Thank you for contributing to the repository.

A friend of mine developed an introductory colab tutorial on Hypergraph Networks using PyTorch 🥳. If you are curious how to process connections beyond pairwise, check out github.com/hypergraphlab/…

📌5 years after revealing the CLAIRE vision for the future of “AI made in Europe”, CLAIRE has released an urgent call for action with a #statement on #AI #Research & #Innovation in #Europe 🔗Read our full statement here!➡️claire-ai.org/ai-made-in-eur… ❓Let us know your thoughts!

With the ELLIS Institute, Cyber Valley is positioning itself to compete globally as the place to be in terms of scientific excellence and inspiring work opportunities. mwk.baden-wuerttemberg.de/de/service/pre… Wissenschaft.Kunst.BW ELLIS

Mathieu Yeah :) There's a super obvious one: learn your favourite routing algorithm, and deploy it in the real world. This recent work from Dobrik Georgiev and @danilo_numeroso is also quite cool, and related: arxiv.org/abs/2306.06064

New blog post! 🔬 An introduction to generative modelling for small molecule drug design with diffusion models 🧬 Generative modelling for molecular design of all kinds has been gaining significant attention recently (1/7) medium.com/@cch57/explori…

🎉 Excited to announce our 2nd KnowledgeNLP Workshop at #KDD2023! 🔍 Join us for an enlightening journey into the world of Knowledge Augmented #NLP. 🌐 Engage with leading experts, explore cutting-edge research, and unlock the power of big data. Don't miss out! #AI #DataScience

Zhangyang (Atlas) Wang "Unveiling the simplicity in Training Graph Neural Networks" #KDD2023

This is the 7th year after my Ph.D. , and now looking back at my grad school, there are 5 things I wish I had known when I started graduate school. 🧵 1/6

Coincidentally, we also have a Arxiv2023 dataset in our new preprint: arxiv.org/abs/2309.16595 GitHub: github.com/TRAIS-Lab/LLM-…

🎉New paper alert! It’s time to announce ULTRA - a pre-trained foundation model for knowledge graph reasoning that works on _any_ graph and outperforms supervised SOTA models on 50+ graphs. Arxiv: arxiv.org/abs/2310.04562 Code: github.com/DeepGraphLearn… 🧵1/n

Mark your calendars -- LoG 2023 is just 10 days away! Schedule available here: logconference.org

The DEGAS webinar series is back. The new season will begin on 6 Dec 3:00pm CET when xiaowen dong present "On the stability of spectral graph filters and beyond: A topological perspective". Check out dsiseminar.github.io for details. #graphsignalprocessing

Checklist of Tips for Computer Science Faculty Applications by Yisong Yue link.medium.com/AmAA8V2CUEb

Still more than a year behind for the Machine Learning Journal (@[email protected]) #ECMLPKDD2023 journal track papers, though: "Generalizing p-Laplacian: spectral hypergraph theory and a partitioning algorithm" by Shota Saito & Mark Herbster (link.springer.com/article/10.100…) #OA

A lot of people are sceptical of Google’s ability to compete in LLM space, but when it comes to time series google claims are turning into some sort of comedy show. A new paper from Google claims “foundational model”, but as it seems to have become a custom from when it comes to

Any algorithms for the shortest path problem which do non-local operations?

Anton Tsitsulin The struggle is real

Check out our new blog post on talking like a graph!