Casper Kerrén

ckerren.bsky.social

20 October 2026 07:30

Imagine two people standing in the same spot, looking at the same Goya. They inspect exactly the same parts of the painting, but in a different order. Later, they talk about what they saw. Ignoring all other differences: are the memories they formed identical?

Qiaoli Huang

qiaoli-huang.bsky.social

17 October 2026 18:57

Very happy to share our new work. Do our eye movements reveal how we structure information in working memory? We found that eye movements adapt to memory demands, shifting from sequential sampling during encoding to replay-like revisitation during maintenance. dx.plos.org/10.1371/jour...

Cheng Xue

cxue.bsky.social

16 October 2026 19:59

How do we remember an image after seeing it just once? 🧠🧪 Now out in @natcomms.nature.com: neurons in mid-level visual cortex (V4) carry signatures that predict successful one-shot recognition memory. Led by Grace DiRisio, w/ @marlenecohen.bsky.social doi.org/10.1038/s41467-026-78369-0

Ann Huang

annhuang42.bsky.social

14 October 2026 19:38

How much representational diversity exists around a trained network? A lot, it turns out! Moving a ViT’s weights by 1.3% while preserving its predictions, its representations end up less similar to the original than a random-init ViT’s ‼️ New paper: Hessian Null Space Continuation (HNC) 💫🧵 (1/12)

Chris Polaris

chrispolaris.com

14 October 2026 19:35

Different internals with matched task behaviour raise a useful next question: do the alternative networks also fail differently under distribution shift? That could turn representational diversity into something operationally valuable.