NSD-synthetic, the out-of-distribution companion dataset of NSD consisting of 7T fMRI responses to 284 artificial images, is now published. #NeuroAI #CompNeuro #neuroscience #AI doi.org/10.1038/s414...

This thread aligns closely with the core claim of our paper. While naturalistic stimuli are highly valuable, large-scale natural data can yield spurious successes due to unintended shortcuts in complex analysis pipelines. doi.org/10.1016/j.ne...

Blake Richards@tyrellturing.bsky.social · 9mo ago

I don't disagree! That being said, I think if you can both use naturalistic stimuli and make manipulations to test hypotheses, that's the ideal situation.

Our paper is now accepted at Neural Networks! This work builds on our previous threads in X, updated with deeper analyses. We revisit brain-to-image reconstruction using NSD + diffusion models—and ask: do they really reconstruct what we perceive? Paper: doi.org/10.1016/j.ne... 🧵1/12

And here’s an experimental podcast-style of paper summary, generated via Notebook LM directed by me! Link: notebooklm.google.com/notebook/9c8...

Ken Shirakawa@kencan7749.bsky.social · last yr.

Our paper is now accepted at Neural Networks! This work builds on our previous threads in X, updated with deeper analyses. We revisit brain-to-image reconstruction using NSD + diffusion models—and ask: do they really reconstruct what we perceive? Paper: doi.org/10.1016/j.ne... 🧵1/12

Our paper is now accepted at Neural Networks! This work builds on our previous threads in X, updated with deeper analyses. We revisit brain-to-image reconstruction using NSD + diffusion models—and ask: do they really reconstruct what we perceive? Paper: doi.org/10.1016/j.ne... 🧵1/12

Yukiyasu Kamitani, Misato Tanaka, Ken Shirakawa Visual Image Reconstruction from Brain Activity via Latent Representation https://arxiv.org/abs/2505.08429

One big issue with some of the previous claims are that NSD, the massive 7T fMRI dataset of 1000s of images, might not be the right dataset to test these hypotheses. The reason is that it is built on MSCoCo and has too high similarity between training and test. arxiv.org/abs/2405.10078 16/n

I’m currently concerned about what the brain’s encoding model predicts. Given that the target brain state is collected under naturalistic condition and the inputs of encoding model derived from a deep neural network, I am not sure what the predictions are actually represent.

arxiv.org/abs/2405.10078

Spurious reconstruction from brain activity

Advances in brain decoding, particularly visual image reconstruction, have sparked discussions about the societal implications and ethical considerations of neurotechnology. As these methods aim to re...

arxiv.org