All news
-
Welcome Daniele and Matt!
Daniele Savietto and Matt Stevenson have joined NPL as PhD students at the Network Science Institute. Welcome to the lab! -
New blindsight preprint out!
Davide's new preprint Preserved brain hierarchy supports residual vision without awareness after visual cortex damage is out on bioRxiv! In the largest resting-state fMRI cohort of patients with unilateral V1 damage reported to date, we show that blindsight-positive patients preserve the sensory-to-association cortical hierarchy, while blindsight-negative patients show a marked flattening of it driven by abnormally low shared-information coupling within unimodal cortices.Read more ↗ -
We are hiring: Postdoc in Topological Machine Learning and Generative AI!
We are looking for a Postdoctoral Research Associate in Topological Machine Learning and Generative AI to lead our contribution to TOPO-LLM, bringing persistent homology and higher-order network methods to bear on large language models for innovation intelligence. Ten months, EUR 64,000, based at the Dublin Innovation Institute - Northeastern University. Applications close at 12pm on 9 September 2026.Read more ↗ -
TOPO-LLM wins FFplus (EuroHPC) funding!
Our proposal TOPO-LLM: Topologically-Informed Large Language Models for Innovation Intelligence has been awarded funding through the FFplus EuroHPC Second Open Call! Gio joins as Scientific Advisor on the project, which brings pre-exascale European HPC to bear on topologically-informed generative AI for innovation intelligence.Read more ↗ -
New sperm whale foraging preprint out!
Our new preprint Sperm whale foraging combines persistent search with flexible prey-capture tactics is out on bioRxiv! Using 34 sensor-tag deployments, we show that sperm-whale foraging is organized across two nested levels: persistent, Levy-like search paths and two distinct, flexibly switched prey-capture tactics.Read more ↗ -
Higher-order brain models paper accepted at Nature Communications!
Our paper Charting Higher-Order Models of Brain Function Beyond Pairwise Interactions is accepted at Nature Communications! Led by Andrea and Matteo, we benchmark higher-order interaction metrics on fMRI data, showing they align with the brain's sensorimotor-to-association hierarchy while dissociating along molecular lines, and that they outperform pairwise connectivity for individual fingerprinting.Read more ↗ -
NPL at OHBM 2026!
Gio, Andrea, and Davide are heading to OHBM 2026 in Bordeaux (June 14–18)! Andrea is giving a talk and Davide is presenting a poster.Read more ↗ -
Gio keynote at Universeh General Meeting!
Giovanni is giving a keynote on The Cost of Sharing at the Universeh General Meeting at the University of Rome Tor Vergata (June 10–12).Read more ↗ -
Jesseba co-organizing Neuromonster 2026!
Jesseba is a co-organizer of Neuromonster, the 7th International Conference on the Mathematics of Neuroscience and AI, happening at Villa Wolkonsky in Rome (June 9–12).Read more ↗ -
NPL at NetSci 2026!
Gio, Henrique, and Jesseba were at NetSci 2026 in Boston! Henrique gave a talk on Social Influence in Large Language Models (LLMs) Groups During Cognitive Tasks, and Gio spoke at the NSIA satellite on The cost of sharing. -
Henrique and Francesco at the British NetSci Symposium!
Henrique and Francesco presented their work at the 3rd British Network Science Society Symposium at King's College London! Henrique spoke on Social Influence in Large Language Models (LLMs) Groups During Cognitive Tasks and Francesco presented Harmonic Morphisms and Dynamical Invariants in Network Renormalization.Read more ↗ -
VLM paper accepted at ICML 2026!
Our paper The Geometry of Representational Failures in Vision Language Models is accepted at ICML 2026! We show that geometric overlap between distilled concept vectors predicts specific multi-object failure patterns in open-weight VLMs.Read more ↗ -
Team heading to ICLR 2026!
Marco, Pierrick, and Gio will be heading to ICLR 2026 to present our work on generalization and topology in neural networks!Read more ↗ -
Future Wunderkammer virtual salon with Berggruen Institute
Gio will be participating in the Future Wunderkammer virtual salon organized by the Berggruen Institute on April 16th at 11am Pacific time. Join via Zoom.Read more ↗ -
Gio speaking at Biennale e Tecnologia!
Giovanni will be speaking at Biennale e Tecnologia on 'Capirsi tra specie: parlare con gli animali e con gli alieni' (Understanding between species: talking with animals and aliens).Read more ↗ -
Harmonic morphisms paper out on arXiv!
New paper on harmonic morphisms and dynamical invariants in network renormalization is now available! We introduce harmonic degree as a tool for assessing coarse-graining quality and show how Laplacian renormalization achieves exact harmonic morphisms in certain networks.Read more ↗ -
Sperm whale birth paper out in Science!
Our paper on cooperative sperm whale birth behavior is published in Science! We document how non-kin cooperation during birth underpins sperm whale social complexity.Read more ↗ -
Companion paper on sperm whale birth out in Scientific Reports!
Our companion paper describing the collaborative sperm whale birth event and shifts in coda vocal styles is out in Scientific Reports!Read more ↗ -
Marco defends his PhD and heads to Princeton!
Huge congratulations to Marco Nurisso, who successfully defended his PhD viva at the Politecnico di Torino on March 25th! Marco will be joining the NAM program at Princeton. -
Marilyn moving on to Assistant Professor at NUL.
Congrats to Marilyn who is starting a position as Assistant Professor at Northeastern University London! -
Room packed for NeuroRenorm workshop at COSYNE 2026!
Great success for the NeuroRenorm workshop on renormalization principles in neural systems at COSYNE 2026! Andrea, Jesseba, and Gio are organizing. -
NeuroRenorm: COSYNE 2026 workshop approved!
Andrea, Jesseba, and Gio will organize a workshop at COSYNE2026 on the renormalization principles in neural systems. Website out soon! -
Two papers accepted at ICLR 2026!
Marco, Jesseba, Raj, Pierrick and Gio got two papers accepted at ICLR 2026: Bound by semanticity and Topology and geometry of the learning space of ReLU networks. -
ABIM 2026
Andrea, Lorenzo, Davide and Gio will be presenting at ABIM 2026. -
Gio at CNWW2025
Gio will be lecturing at the Complex Networks Winter Workshop in Quebec City. -
Jesseba as Accelnet fellow
Jesseba got accepted as a Accelnet Multinet fellow and will be visiting London and Marseille to work with Alain Barrat and Andrea Brovelli in Spring 2026! -
New review on multilayer networks out
New review out reviewing the state of multilayer network science here. -
New work on stability of higher-order systems
New paper out the role of higher-order coupling in stabilizong new states here. -
Gio keynote at the Hypnosis Research Summit
Gio will be attending the Hypnosis Research Summit in Budapest and give a keynote talk on topological descriptions of functional equivalence in mental imagery tasks in sujbects susceptible to hypnosis. -
First RUNES workshop!!
Exploiting a nice synergy, we are organizing the first RUNES workshop in Marseilles, colocated with the BeyondTheEdge first school! -
New paper on PLoS Computational Biology
Our paper led by Marilyn on computational modeling of TUS plasticity effects is out on PLoS Computational Biology. -
Science and Artificial Intelligence Summit 2025
Gio will speak about the tradeoff between identification and generalization at the AI Huawueisummit in London. -
Workshop on topology of human and synthetic minds
Gio is organizing with Yasser Roudi and Federico Turkheimer a workshop on topology for neuroscience at the new King's Institute of Human and Synthetic Minds on Sept 15-16. -
NPL presents topological signal processing research at EUSIPCO 2025
Andrea will present work with Marco and Gio on topological signal processing of brain data at EUSIPCO 2025 in Palermo, Italy (September 8-12). The European Signal Processing Conference is the flagship conference of EURASIP, featuring the latest developments in signal processing research. -
CoBrain at CCS2025
-
Gio attending PIBBSS retreat on renormalization and AI
Gio will be attending a PIBBSS retreat on renormalization and AI from June 27-30. PIBBSS is exploring how renormalization can be developed into a rich framework for AI interpretability. -
New comparative study on higher-order brain connectivity metrics
We present the first large-scale comparison of information-theoretic and topological HOI metrics using fMRI data from the Human Connectome Project. We identify a taxonomy of redundant, synergistic, and topological metrics, with topological ones bridging the two. These metrics align with brain hierarchies, track neurochemical gradients, and outperform pairwise models in fingerprinting and behavior prediction. Paper here, led by Andrea, Simone, Davide, and Marilyn. -
NPL at OHBM 2025
Gio and Andrea will be at OHBM 2025 in Brisbane, Australia. Gio is organizing the HOI (Higher-Order Interactions) symposium with Manish Saggar. They will be presenting the HOI lens and the HOI movies papers. -
New paper out on topological fingerprinting
We show that topological fingerprinting, based on homological scaffolds, significantly outperforms FC-based one. We also show that these scaffolds are distributed across functional subnetworks and we link the structure of topological cycles to information synergy. Paper here led by Simone and Andrea. -
New paper online on the effects of congenital sensory deprivation
We show that while the brain's large-scale functional architecture is shaped by innate hierarchical gradients, sensory deprivation induces targeted, experience-driven reorganization that flexibly reconfigures cortical connectivity without disrupting the brain's core scaffold. Paper here led by Davide. -
Paper on the informational structure of goal-directed learning accepted on Nature Communications!
We show how the brain encodes information gain during goal-directed learning, showing that higher-order synergistic interactions across distributed cortical networks—particularly involving prefrontal circuits—enable the integration and broadcasting of learning signals critical for adaptive decision-making. Paper here led by BraiNet. -
New paper out on the fundamental processing limits in learning systems
We reveal a fundamental informational limit governing the tradeoff between generalization and identification, showing that finite semantic resolution imposes universal constraints on representational capacity across neural networks and brain-like systems. Paper here led by Marco with Jesseba and Raj. -
Jesseba at Santa Fe Institute workshop on Adaptive Networks
Jesseba Fernando is participating in the Foundations of Adaptive Networks working group at the Santa Fe Institute from May 6-10, 2025. The workshop focuses on advancing theoretical foundations of adaptive temporal network models and their applications to complex systems. -
Marilyn as IAS Fellow organizes HOI-BrainMod Workshop
Marilyn Gatica will be a visiting Fellow during spring at the Institute for Advanced Study (IAS) in Amsterdam and will organize the HOI-BrainMod Workshop on Brain Modeling and High-Order Interactions on May 16, 2025. -
New paper out on higher-order renormalization in Nature Physics!
We propose a new Laplacian perspective to renormalize higher-order systems. Paper here. -
SYNAPSE2025
-
Andrea wins the MSCA fellowship!
Andrea has been awarded a MSCA fellowship to work on temporal higher-order approaches to neuropixel datasets. Fun times ahead! More details here -
AATRN seminar
Gio spoke at the AATRN seminar series on recent work bridging mechanisms and behaviours in higher-order systems. Slides are available here. -
NetsciX 2025
Jesseba will present results on the information integration differences between rodents and neural network models in learning-relearning task. Gio will talk as an invited speaker at the Focus Session on Hypergraphs. -
New paper out on computational modeling of TMS stimulation
In a new paper we use human fMRI data and whole-brain computational models to investigate two key mechanisms —-distance-based propagation and diffusion-like broadcasting—- to explain the spreading of the plasticity effect of transcranial ultrasound stimulations -
NeurIPS2024
Marco and Pierrick are going to Vancouver to present their paper . -
ERC Consolidator Grant RUNES!
Gio is awarded the Consolidator Grant by the European Research Commission for the proposal RUNES: Reconstruction and unification of neural and ecological systems. More info available here. -
New paper on Nature Communications out!
Andrea and Gio published a paper titled Higher-order connectomics of human brain function reveals local topological signatures of task decoding, individual identification, and behavior in Nature Communications on the topological structure of instantaneous higher-order brain activations across tasks and individuals. The paper is available here. -
Statistical Mechanics of Cognition
Gio will be talking at the new workshop coming up at the end of the November. Program is available here . -
IPAM Mathematics of Intelligence
Simone and Henrique accepted to IPAM's her collective intelligence workshop at UCLA. -
BrainModes 2024
Marilyn talking at BrainModes Conference 2024 in Bilbao. -
Project CETI all-team meeting in Dominica
Gio, Simone, and Andrea will be participating in the 2024 Project CETI all-team meeting from 7 to 11th November at the Dominica HQ. -
KIAS-KU workshop on Theoretical Challenges in Network Science
Gio will be talking at the KIAS-KU workshop on Theoretical Challenges in Network Science at the beginning of the November. Program is available here . -
New NPL website launched!
After a looooooong time, our new website is finally out! It is still in construction, but the main functionalities are there. -
New NPC!
Andrea Santoro joins the lab as CETI postdoc! -
New PRX paper out!
'Higher-Order Null Models as a Lens for Social Systems' now out on PRX here. -
Two new NPL members!
Henrique M. Borges and Raj Deshpande join the lab as NetSI PhD students. -
New NPL member!
Marilyn Gatica joins the lab! -
NPL grows!
Gio joins Northeastern University London as a Professor in the Network Science Institute. -
Between mechanisms and observables in higher-order systems
Link to slides of Giovanni Petri's talk at the conference: Navigating the Research Frontier of AI and Complexity here.