NIPS-CERN is happy to share that two papers co-authored by members of the group were accepted for presentation at the Brazilian Symposium on Telecommunications and Signal Processing (SBrT 2026), which runs in Salvador, Bahia, from 29 September to 2 October 2026. Both continue the group's work on signal processing in calorimetry and machine learning for the ATLAS experiment at the LHC/CERN.

• Optimization strategies for CNNs aiming crosstalk mitigation in finely segmented calorimetry systems. A collaboration between Brazil and France that brings together UFBA, UFJF and the LPNHE (Sorbonne Université / CNRS, in Paris), by André Paiva Conrado Rodrigues, Antonio Carlos Lopes Fernandes Junior, Paulo Farias, Eduardo Simas Filho, Edmar Egídio Purcino de Souza, Mateus H. M. Faria (a NIPS-CERN doctoral student), Luciano Manhães de Andrade Filho and Bertrand Laforge.
• Convolutional Neural Networks for Signal Estimation in High-Energy Calorimeters, written with IPRJ/UERJ by Diogo Alves Cardinot, Bernardo Sotto-Maior Peralva, Gustavo Barbosa Libotte and Luciano Manhães de Andrade Filho. It carries on the group's work on neural-network signal reconstruction in calorimetry, the same line as its recent paper in Applied Sciences.

The convolutional network in that second paper is the one undergraduate Clethie Sediacias Bisseyou Ngatsono is implementing on the SAPHO processor for his final-year project, co-advised by doctoral student Leandro da Graça Ribeiro. It connects the collaboration with UERJ to work already under way in the group.

Congratulations to everyone involved.

Cover photo: Pelourinho, the historic centre of Salvador. Photograph by Paul R. Burley, CC BY-SA 4.0, via Wikimedia Commons.