Beatrice Bevilacqua

Bio

I am a Machine Learning Research Scientist at the Arc Institute, where I develop and evaluate large-scale models that predict how cells respond to perturbations. My work spans the full model lifecycle, from training on observational data and post-training on perturbational data to rigorous evaluation across biological contexts.

These challenges connect to a central theme of my research: building models that generalize beyond their training distributions. My earlier work studied how symmetries and causal structure can support generalization across domains. I received my Ph.D. in Computer Science from Purdue University, advised by Prof. Bruno Ribeiro and in close collaboration with Prof. Haggai Maron, and previously interned at Google DeepMind and Meta FAIR.

Experience

• Arc Institute May 2025 - Present

  ML Research Scientist
  End-to-end model development, post-training, and evaluation for cellular response prediction
• Google DeepMind Sep 2022 - Jan 2023

  Research Scientist, Intern
  Synthetic data and causal learning for neural algorithmic reasoning
• Meta AI (FAIR) May 2022 - Aug 2022

  Research Scientist, Intern
  Causal self-supervision for image models under domain shift
• Purdue University 2020 - 2025

  Ph.D.
  Computer Science

Publications

Most recent publications on Google Scholar. ^‡ indicates equal contribution.

• Selected
• All

Predicting cellular responses to perturbation across diverse contexts with State

Abhinav K. Adduri, Dhruv Gautam, Beatrice Bevilacqua, Alishba Imran, Rohan Shah, Mohsen Naghipourfar, Noam Teyssier, Rajesh Ilango, Sanjay Nagaraj, Mingze Dong, Chiara Ricci-Tam, Christopher Carpenter, Vishvak Subramanyam, Aidan Winters, Sravya Tirukkovular, Jeremy Sullivan, Brian S. Plosky, Basak Eraslan, Nicholas D. Youngblut, Jure Leskovec, Luke A. Gilbert, Silvana Konermann, Patrick D. Hsu, Alexander Dobin, Dave P. Burke, Hani Goodarzi, Yusuf H. Roohani

Cell 2026

Paper

Bridging Input Feature Spaces Towards Graph Foundation Models

Moshe Eliasof, Krishna Sri Ipsit Mantri, Beatrice Bevilacqua, Bruno Ribeiro, Carola-Bibiane Schönlieb

ICLR 2026

Paper

Holographic Node Representations: Pre-training Task-Agnostic Node Embeddings

Beatrice Bevilacqua, Joshua Robinson, Jure Leskovec, Bruno Ribeiro

ICLR 2025

Paper Code

Granola: Adaptive Normalization for Graph Neural Networks

Moshe Eliasof^‡, Beatrice Bevilacqua^‡, Carola-Bibiane Schönlieb, Haggai Maron

NeurIPS 2024

Paper

Neural Algorithmic Reasoning with Causal Regularisation

Beatrice Bevilacqua, Kyriacos Nikiforou, Borja Ibarz, Ioana Bica, Michela Paganini, Charles Blundell, Jovana Mitrovic, Petar Veličković

ICML 2023

Paper

Understanding and Extending Subgraph GNNs by Rethinking Their Symmetries

Fabrizio Frasca^‡, Beatrice Bevilacqua^‡, Michael M. Bronstein, Haggai Maron

NeurIPS 2022, Oral (top 1.9% out of 9,634 submissions)

Paper Code

Equivariant Subgraph Aggregation Networks

Beatrice Bevilacqua^‡, Fabrizio Frasca^‡, Derek Lim^‡, Balasubramaniam Srinivasan, Chen Cai, Gopinath Balamurugan, Michael M. Bronstein, Haggai Maron

ICLR 2022, Spotlight (top 5.1% out of 3,391 submissions)

Paper Video Code

Predicting cellular responses to perturbation across diverse contexts with State

Abhinav K. Adduri, Dhruv Gautam, Beatrice Bevilacqua, Alishba Imran, Rohan Shah, Mohsen Naghipourfar, Noam Teyssier, Rajesh Ilango, Sanjay Nagaraj, Mingze Dong, Chiara Ricci-Tam, Christopher Carpenter, Vishvak Subramanyam, Aidan Winters, Sravya Tirukkovular, Jeremy Sullivan, Brian S. Plosky, Basak Eraslan, Nicholas D. Youngblut, Jure Leskovec, Luke A. Gilbert, Silvana Konermann, Patrick D. Hsu, Alexander Dobin, Dave P. Burke, Hani Goodarzi, Yusuf H. Roohani

Cell 2026

Paper

Bridging Input Feature Spaces Towards Graph Foundation Models

Moshe Eliasof, Krishna Sri Ipsit Mantri, Beatrice Bevilacqua, Bruno Ribeiro, Carola-Bibiane Schönlieb

ICLR 2026

Paper

Zero-Shot Generalization of GNNs over Distinct Attribute Domains

Yangyi Shen, Jincheng Zhou, Beatrice Bevilacqua, Joshua Robinson, Charilaos Kanatsoulis, Jure Leskovec, Bruno Ribeiro

ICML 2025

Paper

Holographic Node Representations: Pre-training Task-Agnostic Node Embeddings

Beatrice Bevilacqua, Joshua Robinson, Jure Leskovec, Bruno Ribeiro

ICLR 2025

Paper Code

TRIX: A More Expressive Model for Zero-shot Domain Transfer in Knowledge Graphs

Yucheng Zhang, Beatrice Bevilacqua, Mikhail Galkin, Bruno Ribeiro

LoG 2024

Paper

Digraf: Diffeomorphic Graph-Adaptive Activation Function

Krishna Sri Ipsit Mantri, Xinzhi Wang, Carola-Bibiane Schönlieb, Bruno Ribeiro, Beatrice Bevilacqua^‡, Moshe Eliasof^‡

NeurIPS 2024

Paper

Granola: Adaptive Normalization for Graph Neural Networks

Moshe Eliasof^‡, Beatrice Bevilacqua^‡, Carola-Bibiane Schönlieb, Haggai Maron

NeurIPS 2024

Paper

Subgraphormer: Unifying Subgraph GNNs and Graph Transformers via Graph Products

Guy Bar-Shalom, Beatrice Bevilacqua, Haggai Maron

ICML 2024

Paper

Efficient Subgraph GNNs by Learning Effective Selection Policies

Beatrice Bevilacqua, Moshe Eliasof, Eli Meirom, Bruno Ribeiro, Haggai Maron

ICLR 2024

Paper Code

A Multi-Task Perspective for Link Prediction with New Relation Types and Nodes

Jincheng Zhou, Beatrice Bevilacqua, Bruno Ribeiro

NeurIPS GLFrontiers 2023

Paper

Neural Algorithmic Reasoning with Causal Regularisation

Beatrice Bevilacqua, Kyriacos Nikiforou, Borja Ibarz, Ioana Bica, Michela Paganini, Charles Blundell, Jovana Mitrovic, Petar Veličković

ICML 2023

Paper

Graph Positional Encoding via Random Feature Propagation

Moshe Eliasof, Fabrizio Frasca, Beatrice Bevilacqua, Eran Treister, Gal Chechik, Haggai Maron

ICML 2023

Paper

Causal Lifting and Link Prediction

Leonardo Cotta, Beatrice Bevilacqua, Nesreen Ahmed, Bruno Ribeiro

The Royal Society A

Paper

A Generalist Neural Algorithmic Learner

Borja Ibarz, Vitaly Kurin, George Papamakarios, Kyriacos Nikiforou, Mehdi Bennani, Róbert Csordás, Andrew Dudzik, Matko Bošnjak, Alex Vitvitskyi, Yulia Rubanova, Andreea Deac, Beatrice Bevilacqua, Yaroslav Ganin, Charles Blundell, Petar Veličković

LoG 2022, Spotlight

Paper

Understanding and Extending Subgraph GNNs by Rethinking Their Symmetries

Fabrizio Frasca^‡, Beatrice Bevilacqua^‡, Michael M. Bronstein, Haggai Maron

NeurIPS 2022, Oral (top 1.9% out of 9,634 submissions)

Paper Code

Equivariant Subgraph Aggregation Networks

Beatrice Bevilacqua^‡, Fabrizio Frasca^‡, Derek Lim^‡, Balasubramaniam Srinivasan, Chen Cai, Gopinath Balamurugan, Michael M. Bronstein, Haggai Maron

ICLR 2022, Spotlight (top 5.1% out of 3,391 submissions)

Paper Video Code

Size-Invariant Graph Representations for Graph Classification Extrapolations

Beatrice Bevilacqua^‡, Yangze Zhou^‡, Bruno Ribeiro

ICML 2021, Long Talk (top 3% out of 5,513 submissions)

Paper Code

Simulation-Based Evolutionary Optimization of Air Traffic Management

Alessandro Pellegrini, Pierangelo di Sanzo, Beatrice Bevilacqua, Gabriella Duca, Domenico Pascarella, Roberto Palumbo, Juan Josè Ramos, Miquel Àngel Piera, Gabriella Gigante

IEEE Access 2020

Service

• Instructor of Set and Graph Representations Classes within CS58700 ``Foundations of Deep Learning'' (Spring 2022 & Spring 2023), held by Prof. Bruno Ribeiro
• Reviewer: ICLR, ICML, and NeurIPS (2022–2026)
• Top Reviewer NeurIPS 2022, 2023, 2024, 2025
• Reviewer TPAMI 2023, 2025
• Reviewer TMLR 2024, 2025

Talks and Tutorials

• Towards Graph Foundation Models
  GeorgiaTech (virtual), Nov 2024
• Scalability of Subgraph GNNs [Link]
  GeorgiaTech (virtual), Feb 2024
• Double Exchangeability in Knowledge Graphs Endows Neural Networks with Logical Knowledge Transfer [Link]
  NetSci 2023, July 2023
  Lightning Talk Speaker in a plenary session
• Subgraphs for Enhanced Graph Neural Networks [Link]
  Johns Hopkins University, Feb 2023
• Exploring the practical and theoretical landscape of expressive Graph Neural Networks [Link]
  LOG 2022 (virtual), Dec 2022
  Tutorial Organizer and Speaker
• Understanding and extending subgraph GNNs by rethinking their symmetries [Link]
  Learning on Graphs and Geometry Reading Group (virtual), Nov 2022
• Pack your subgraphs: A journey into subgraphs for powerful Graph Neural Networks (GNNs) [Link]
  Sapienza University of Rome (virtual), Nov 2022
• Pack your subgraphs: A journey into subgraphs for powerful Graph Neural Networks
  DeepMind Reading Group (London), Sept 2022
• Pack your subgraphs: A journey into subgraphs for powerful Graph Neural Networks
  Meta AI orgs Reading Meeting (Menlo Park), Aug 2022
• Pack your subgraphs: A journey into subgraphs for powerful Graph Neural Networks
  École Polytechnique (virtual), July 2022
• Equivariant Subgraph Aggregation Networks [Link]
  WelcomeAIOverlords (virtual), March 2022
• Equivariant Subgraph Aggregation Networks [Link]
  Learning on Graphs and Geometry Reading Group (virtual), Dec 2021