(1/3) The Algonauts session at CCN has concluded: thank you to all who participated! The session recording will soon be available on the YouTube CCN page.

#algonauts2025 #NeuroAI #CompNeuro #neuroscience #AI #CCN2025 #hessianAI #CNeuroMod #FUBerlin #GoetheUni

(1/2) The Algonauts Project 2025 challenge results are out! 1st: sdascoli; 2nd: NCG; 3rd: SDA. Congratulations! Join the Algonauts session at CCN to know how they did it.

#algonauts2025 #NeuroAI #CompNeuro #neuroscience #AI #CCN2025 #hessianAI #CNeuroMod #FUBerlin #GoetheUni

2025.ccneuro.org

The Algonauts Challenge submission is over: > 60 teams participated! To be considered in the winners selection process, please post your report/code by July 25th. We will announce the winners on August 5th!

#NeuroAI #CompNeuro #neuroscience #AI #CCN2025 #hessianAI #CNeuroMod

algonautsproject.com

On July 6th will open the final Algonauts challenge phase, which will test the out-of-distribution generalization of your models, and determine the challenge winners.

We look forward to your submissions!

algonautsproject.com

Vision Deep Neural Networks and Large Language Models complement each other in predicting EEG responses to images, providing a more accurate model of the time course of visuo-semantic processing in the human brain.

#NeuroAI #CompNeuro #neuroscience #AI

arxiv.org/abs/2506.19497

Using in silico neuroscience, @alessandrogifford.bsky.social et al. developed a neural control algorithm to modulate the representational relationships between visual cortical areas, revealing how these areas jointly represent the world as an interconnected network.
www.nature.com/articles/s41...

Update: hessian.AI will sponsor up to 1,600€ for travel expenses and conference registration fees for each of the top three Algonauts 2025 Challenge teams to present their winning models at CCN 2025!

@hessianai.bsky.social @cogcompneuro.bsky.social

algonautsproject.com

(7/7) NSD-synthetic is freely available at:

naturalscenesdataset.org

(6/7) Out-of-distribution generalization tests on NSD-synthetic reveal differences between brain models that are not detected by in-distribution tests => self-supervised deep neural networks better explain neural responses than their task-supervised counterparts.

(5/7) NSD-synthetic is a stricter testbed that highlights deficiencies of brain encoding models not observable with in-distribution tests.

(4/7) NSD-synthetic’s fMRI responses are out-of-distribution with respect to NSD-core.

(3/7) NSD-synthetic’s fMRI responses reliably encode stimulus-related information over visual cortex.

(2/7) NSD-synthetic consists of 7T fMRI responses from the eight NSD subjects for 284 carefully controlled synthetic images, divided into 8 classes.

(1/7) NSD-synthetic is out!

NSD-synthetic enables strict out-of-distribution generalization tests critical for the development of more robust models of visual processing and the formulation of more accurate theories of human vision.

#NeuroAI #CompNeuro #neuroscience #AI

doi.org/10.48550/arX...

(1/4) The Algonauts Project 2025 challenge is now live!

Participate and build computational models that best predict how the human brain responds to multimodal movies!

Submission deadline: 13th of July.

#algonauts2025 #NeuroAI #CompNeuro #neuroscience #AI

algonautsproject.com