Giuseppe Costantino

Postdoctoral researcher, École Normale Supérieure, Paris @ENS_ULM | artificial intelligence (AI), geophysics | AI-empowered observation of the seismic cycle

[Shunsuke Takemura（武村俊介）🌏 @takeshun1984]

Costantino, G., & Jolivet, R. (2026). Denoising of InSAR time series through spatiotemporal attentive convolutional U-net. Journal of Geophysical Research: Solid Earth, 131, e2026JB033940. https://t.co/E2iYQuWRVQ

[Shunsuke Takemura（武村俊介）🌏 @takeshun1984]

Wech, A. G., & Gomberg, J. (2025). Rupture continuity through intermittent pauses in Cascadia slow slip events. Journal of Geophysical Research: Solid Earth, 130, e2025JB031501. https://t.co/lQ0KBkkyai

[Nature Geoscience @NatureGeosci]

Article: Ruptures on more structurally mature faults are more localized and therefore expected to host faster ruptures with less off-fault deformation

https://t.co/keJcIMFn2d

[Earth Planets Space @eps_tweet]

Express Letter on Seismology
Y. Jinde et al.(Published: 22 July 2025)
Enhanced deep learning approach for detecting and locating tectonic tremors in the Nankai subduction zone
https://t.co/CAg1VBVJTi

[Dr. Alice-Agnes Gabriel @InSeismoland]

A digital twin for Cascadia tsunami early warning: We present real-time Bayesian inference of seafloor motion at extreme scale, using 3D coupled acoustic-gravity wave equations & El Capitan's 43,520 AMD #GPUs preprint led by @StefanHenneking & Omar Ghattas
https://t.co/N0p5yYf124

[Krittanon (Pond) Sirorattanakul @seismopond]

A chapter from my @Caltech PhD thesis has been published in @Nature. We (me, @StacyLarocheIIe, V. Rubino, N. Lapusta, A. Rosakis) show that interfaces under non-zero shear stress are always sliding, even if they appear to be stationary to the naked eye.

https://t.co/4BHrz15dOQ

[Andy Keller @t_andy_keller]

In the physical world, almost all information is transmitted through traveling waves -- why should it be any different in your neural network?

Super excited to share recent work with the brilliant @mozesjacobs: "Traveling Waves Integrate Spatial Information Through Time"
1/14

[Giuseppe Costantino @giuseppecost95]

[new paper alert!] For the first time, deep learning denoises GNSS time series, revealing hiddent transients of aseismic deformation. The resemblance to tectonic tremor is astonishing! Check it out here: https://t.co/IbZkWSTBbY
#AI #Geoscience #RemoteSensing

[Shunsuke Takemura（武村俊介）🌏 @takeshun1984]

Bombardier, M., Cassidy, J. F., Dosso, S. E., & Kao, H. (2024). Spatial distribution of tremor episodes from long-term monitoring in the northern Cascadia subduction zone. Journal of Geophysical Research: Solid Earth, 129, e2024JB029159. https://t.co/fiq9HJ3ftB

[Seismica @WeAreSeismica]

How big was the biggest earthquake ever recorded in the Campi Flegrei caldera?

Supino et al. shed light on observed ground shaking and help understand possible future scenarios of larger events.

Read more: https://t.co/gNB6rhEgOs

[Seismica @WeAreSeismica]

Seismica Volume 3 No 1 is out now!

Thanks to our authors, reviewers, and volunteers that made this issue possible. These articles and reports present new research and insights from our worldwide community.

📸: Guy Salomon
Full issue: https://t.co/sx88apOhyz

[yosuke aoki @yoaoki]

Slow rupture in a fluid-rich fault zone initiated the 2024 Mw 7.5 Noto earthquake | Science https://t.co/aDjqJHBLJE

[Giuseppe Costantino @giuseppecost95]

@jbourcier10 @valeriomarsocci @IEEE_GRSS Same here 🥲

[Seismica @WeAreSeismica]

What causes onshore faults to slip in subduction zones?

Well, it turns out that the commonly held reason does not apply to the faults located in northern Cascadia.

Read Harrichhausen et al. to find out why: https://t.co/w0tSxMum8P

[Asma Ghandeharioun @ghandeharioun]

🧵Can we “ask” an LLM to “translate” its own hidden representations into natural language? We propose 🩺Patchscopes, a new framework for decoding specific information from a representation by “patching” it into a separate inference pass, independently of its original context. 1/9

[Giuseppe Costantino @giuseppecost95]

Enhanced SSEdetector (https://t.co/oQsz547Ve4) now tracks spatiotemporal tremor evolution in Cascadia using a multi-station approach (blue). We compare it with Xue and Freymueller's (2023) single-station method (red). Check it out at #AGU23! @theAGU
https://t.co/g0rT0IIpgM

[Rikuto Fukushima @rikuto_fksm]

My first paper is now available online!

We applied the Phisycs-Informed Neural Networks (PINNs) to earthquake cycle simulation, and estimated the frictional parameters from SSEs in a spring-slider system.

Please check it out!
https://t.co/NffMqeLedb

[Dr. Deepa Mele Veedu @deepameleveedu]

Multi-station deep learning on geodetic time series detects slow slip events in Cascadia (Costantino et al., 2023)

https://t.co/EVGlAUOPjg

[Giuseppe Costantino @giuseppecost95]

Can deep learning help us detecting transient aseismic deformation hidden in the noise?
Thanks to SSEdetector, we reveal slow slip events in Cascadia using raw GPS data!
Check our new paper out at @CommsEarth @SpringerNature : https://t.co/NyoZOjgExV

[Anne Socquet @AnneSocquet]

In this paper, @giuseppecost95 shows that the temporal evolution of slow slip events retrieved by his deep learning detector is very similar to that of tremor in Cascadia. Check this work out! https://t.co/OJcOkUpclm