@flaNick94
Joined January 2025
  • Tweets7
  • Following33
  • Followers24
1Media
We’re happy to share our latest work on diffusion models, now out as an arXiv preprint. During score learning, different classes in a dataset are generalized and memorized at different times. What drives these differences? Find out here: arxiv.org/abs/2605.06367

The Interplay of Data Structure and Imbalance in the Learning...

Real-world datasets are inherently heterogeneous, yet how per-class structural differences and sampling imbalance shape the training dynamics of diffusion models-and potentially exacerbate...

arxiv.org
1
6
1
35
14,963
@the_chma Enrico Ventura @saglietti_luca @stefsmlab
1
248
Flavio Nicoletti retweeted
Stefano Sarao Mannelli@stefsmlabFeb 18
📢 We’re now accepting applications for the 2026 School on Analytical Connectionism dedicated this year to Language Acquisition. 📍 Gothenburg, Sweden 🗓️ August 17–28, 2026 ☠️ Apply by April 17! 🔗 analytical-connectionism.net… 👇 Meet the experts joining us this summer!
1
15
2
41
10,405
📢 Apply to our workshop on Mathematical Foundations of AI on: Diffusion Models, Transformers and Associative Memories. 📍 Gothenburg, Sweden 🗓️June 8–10, 2026 💀 Register by May 15! We are also accepting contributions! 📎 sites.google.com/view/chair-… 👇 Find our speakers!
1
5
1
35
5,726
speakers: Beatrice Achilli, @LucaAmb, Freya Behrens, @albertofachechi, Alessio Giorlandino, Giovanni Marchetti, Matteo Negri, Daniel Persson, Peter Súkeník, Yedi Zhang
2
206
Flavio Nicoletti retweeted
Stefano Sarao Mannelli@stefsmlabFeb 18
Local Organiser: @flaNick94 School Directors: @ermgrant, me Advisors: @s_y_chung, @SaxeLab
1
2
183
Flavio Nicoletti retweeted
Physical Review B@PhysRevB1 Feb 2025
PRB Editors' Suggestion: #SoftModes in vector #SpinGlass models on sparse random graphs S. Franz, C. Lupo et al., Phys. Rev. B 111, 014203 ➡️ go.aps.org/42BoO8i #EdSugg @APSPhysics #condmat #physics @flaNick94 @wolphysics @FRicciTersenghi @giorgioparisi @smfn_sapienza
EDITORS' SUGGESTION Soft modes in vector spin glass models on sparse random graphs 23 January, 2025 The authors study here low-energy linear excitations in XY and Heisenberg spin glasses at zero temperature, considering systems defined on sparse random regular graphs. The localization properties of the soft modes are linked to the spin glass transition in a field. The authors discover a new kind of low-energy excitation, called a “Concentrated and Delocalized Low Energy Mode” (CDLEM), given that this excitation is restricted to a sub-extensive number of vertices spread over the largest possible distances. Silvio Franz et al. Phys. Rev. B 111, 014203 (2025)

ALT EDITORS' SUGGESTION Soft modes in vector spin glass models on sparse random graphs 23 January, 2025 The authors study here low-energy linear excitations in XY and Heisenberg spin glasses at zero temperature, considering systems defined on sparse random regular graphs. The localization properties of the soft modes are linked to the spin glass transition in a field. The authors discover a new kind of low-energy excitation, called a “Concentrated and Delocalized Low Energy Mode” (CDLEM), given that this excitation is restricted to a sub-extensive number of vertices spread over the largest possible distances. Silvio Franz et al. Phys. Rev. B 111, 014203 (2025)

1
1
12
819
Load more