papers

Publications (15)

physics.flu-dyn2024

Chaotic mixing in plane Couette turbulence

John R. Elton, Predrag Cvitanović, Jonathan Halcrow +1

Lagrangian tracer particle trajectories for invariant solutions of the Navier-Stokes equations confined to the three-dimensional geometry of plane Couette flow are studied. Treatin…

cs.LG2023

Stars: Tera-Scale Graph Building for Clustering and Graph Learning

CJ Carey, Jonathan Halcrow, Rajesh Jayaram +3

A fundamental procedure in the analysis of massive datasets is the construction of similarity graphs. Such graphs play a key role for many downstream tasks, including clustering, c…

cs.LG2023

HUGE: Huge Unsupervised Graph Embeddings with TPUs

Brandon Mayer, Anton Tsitsulin, Hendrik Fichtenberger +2

Graphs are a representation of structured data that captures the relationships between sets of objects. With the ubiquity of available network data, there is increasing industrial…

physics.flu-dyn2008

Heteroclinic connections in plane Couette flow

Jonathan Halcrow, John F. Gibson, Predrag Cvitanović +1

Plane Couette flow transitions to turbulence for Re~325 even though the laminar solution with a linear profile is linearly stable for all Re (Reynolds number). One starting point f…

cs.CL2024

Test of Time: A Benchmark for Evaluating LLMs on Temporal Reasoning

Bahare Fatemi, Mehran Kazemi, Anton Tsitsulin +6

Large language models (LLMs) have showcased remarkable reasoning capabilities, yet they remain susceptible to errors, particularly in temporal reasoning tasks involving complex tem…

cs.LG2024

Let Your Graph Do the Talking: Encoding Structured Data for LLMs

Bryan Perozzi, Bahare Fatemi, Dustin Zelle +4

How can we best encode structured data into sequential form for use in large language models (LLMs)? In this work, we introduce a parameter-efficient method to explicitly represent…

cs.CL2025

Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities

Gheorghe Comanici, Eric Bieber, Mike Schaekermann +3431

In this report, we introduce the Gemini 2.X model family: Gemini 2.5 Pro and Gemini 2.5 Flash, as well as our earlier Gemini 2.0 Flash and Flash-Lite models. Gemini 2.5 Pro is our…

physics.flu-dyn2009

Equilibrium and traveling-wave solutions of plane Couette flow

John F. Gibson, Jonathan Halcrow, Predrag Cvitanović

We present ten new equilibrium solutions to plane Couette flow in small periodic cells at low Reynolds number (Re) and two new traveling-wave solutions. The solutions are continued…

cs.LG2024

Understanding Transformer Reasoning Capabilities via Graph Algorithms

Clayton Sanford, Bahare Fatemi, Ethan Hall +5

Which transformer scaling regimes are able to perfectly solve different classes of algorithmic problems? While tremendous empirical advances have been attained by transformer-based…

cs.LG2020

Grale: Designing Networks for Graph Learning

Jonathan Halcrow, Alexandru Moşoi, Sam Ruth +1

How can we find the right graph for semi-supervised learning? In real world applications, the choice of which edges to use for computation is the first step in any graph learning p…

cs.DC2025

Large-Scale Graph Building in Dynamic Environments: Low Latency and High Quality

Filipe Miguel Gonçalves de Almeida, CJ Carey, Hendrik Fichtenberger +8

Learning and constructing large-scale graphs has attracted attention in recent decades, resulting in a rich literature that introduced various systems, tools, and algorithms. Grale…

cs.LG2023

TF-GNN: Graph Neural Networks in TensorFlow

Oleksandr Ferludin, Arno Eigenwillig, Martin Blais +24

TensorFlow-GNN (TF-GNN) is a scalable library for Graph Neural Networks in TensorFlow. It is designed from the bottom up to support the kinds of rich heterogeneous graph data that…

cs.LG2023

UGSL: A Unified Framework for Benchmarking Graph Structure Learning

Bahare Fatemi, Sami Abu-El-Haija, Anton Tsitsulin +5

Graph neural networks (GNNs) demonstrate outstanding performance in a broad range of applications. While the majority of GNN applications assume that a graph structure is given, so…

cs.LG2023

Talk like a Graph: Encoding Graphs for Large Language Models

Bahare Fatemi, Jonathan Halcrow, Bryan Perozzi

Graphs are a powerful tool for representing and analyzing complex relationships in real-world applications such as social networks, recommender systems, and computational finance.…

physics.flu-dyn2008

Visualizing the geometry of state space in plane Couette flow

John F. Gibson, Jonathan Halcrow, Predrag Cvitanović

Motivated by recent experimental and numerical studies of coherent structures in wall-bounded shear flows, we initiate a systematic exploration of the hierarchy of unstable invaria…