For all of your edgy machine learning needs!

Announcing Graph Mining Atlanta! I'm hiring for a new branch of my team (part of Google Research) in Atlanta this summer. To see more about what we do see our talks from NeurIPS 2020 gm-neurips-2020.github.io and our team page research.google/teams/graph-mi… linkedin.com/feed/update/ur…

🔥Internship opportunity (fall/summer'23) in the Graph Mining team @ Google! Join us to work on cutting edge research in graph ML 🤖. 🧠Prev interns🧠 published in NeurIPS, KDD, WWW, and more! Referral bonus: one (1) cookie 🍪. #ml #internship

2) UGSL: A Unified Framework for Benchmarking Graph Structure Learning (openreview.net/forum?id=9klGk…) Friday 11:10-12pm TLDR: we propose a benchmarking strategy for graph structure learning using a unified framework. Joint work with Anton Tsitsulin Mehran Kazemi Jonathan Halcrow Bryan Perozzi.

Heading out to ICML! Catch me Sunday at the Expo and come learn about TF-GNN icml.cc/Expo/Conferenc… Check this thread for some other work by my group:

🏝 Aloha and good morning! 🏝 #ICML2023 is shaping up to be the most exciting #MachineLearning conference yet, and I'm proud that our graph neural networks group at #Google is participating so much throughout the week! Here's a brief overview of where you can find us:

How can you tell if unsupervised learning "worked or not"? People ask me this all the time! Our TAG-ML paper at #ICML2023 investigates this question - arxiv.org/pdf/2305.16562…

Aloha! 🤙🌺🌴 Check out our 3 papers today at TAGinDS! (11AM p.) Learning Large Graph Property Prediction via Graph Segment Training (11AM p.) UGSL: A Unified Framework for Benchmarking Graph Structure Learning (2.30PM oral + 4PM p.) Unsupervised Embedding Quality Evaluation

Check out latest and greatest in data selection with graphs! The paper features: 🤯 A hardness theorem for training data selection 🕸️ A neural graph approach to solve the problem 🌶️ Comparisons with 10+ SotA methods No graph? No problem, just build one 🏗️ — I promise it works!

Packed room for TFGNN tutorial today at #KDD2023!

FYI: we're running a Kaggle contest with GNNs!

New paper alert 🚨: "Talk Like a Graph" studies the problem of encoding graph-structured data, paving the way for AI to better understand and process complex relationships. Surprising results below 🤯! #AI #LLM #GraphData #MachineLearning

𝐋𝐞𝐭 𝐘𝐨𝐮𝐫 𝐆𝐫𝐚𝐩𝐡 𝐃𝐨 𝐭𝐡𝐞 𝐓𝐚𝐥𝐤𝐢𝐧𝐠: 𝐄𝐧𝐜𝐨𝐝𝐢𝐧𝐠 𝐒𝐭𝐫𝐮𝐜𝐭𝐮𝐫𝐞𝐝 𝐃𝐚𝐭𝐚 𝐟𝐨𝐫 𝐋𝐋𝐌𝐬 Don't know what to do with your graphs in 2024? Shove them to an LLM, of course, and let LLM figure out what to do! arxiv.org/abs/2402.05862 Short thread (1/5):

Talk Like a Graph or Let Your Graph Do the Talking 🧠🕸️ New paper alert 🚨: we show how GNNs can effectively encode graph-structured data for LLMs: arxiv.org/abs/2402.05862

We'll be demoing some of our work on graphs and LLMs later today at ICLR. If you are around stop by and say Hey!

↪️ It's "Talk Like a Graph" day at #iclr2024 - 3:15, at #google see a *live demo* of GraphToken, where we train a tiny model (Gemma 2B) and get sota! arxiv.org/abs/2402.05862 - 4:30, stop by poster 89 to see the TLG leaderboard and talk with us! research.google/blog/talk-like…

🚨 Transformer theory alert 🚨 What algorithms can transformers execute efficiently? Our preprint sheds some light on reasoning capabilities of transformers, now in 𝒓𝒆𝒂𝒍𝒊𝒔𝒕𝒊𝒄 𝒑𝒂𝒓𝒂𝒎𝒆𝒕𝒆𝒓 𝒓𝒆𝒈𝒊𝒎𝒆𝒔. Paper: arxiv.org/abs/2405.18512 More in thread! 1/8

Think your LLM is a time wizard? 🧙 One might think they answer temporal questions through reasoning, but under the hood they may solely rely on their parametric knowledge 🤯 Our new benchmark, ToT, is specifically designed to measure the temporal reasoning capabilities of LLMs.

While I'm at home with my newborn (👶🏻❤️), you can still catch my work at #NeurIPS🧵:

Think your LLM is smart? Try Big-Bench EXTRA Hard. The real test. #Reasoning #Benchmark #LLM