Anastasia Christakou 💙

@christakou.bsky.social

Professor of Neuroscience 👩‍🎓 own views 🐙 anastasia.christakou.org https://orcid.org/0000-0002-4267-3436 children's welfare | science | 🧠 brains & their people

The obvious upshot of recent improvements in LLMs is that things we always knew were easy (theoretical physics, Fields-level mathematics, engineering and computer programming) are, in fact, easy, while the things we knew were hard (writing 8000 words on one sentence from David Lewis) remain hard

To the surprise of no-one who considered the issue for more than a few minutes, "Scientists using LLMs will ‘do more, less well’ modelling study predicts". The use of "less refined", a charitable euphemism lifting mountains in this context... 🤔 www.nature.com/articles/d41...

Scientists using LLMs will ‘do more, less well’, modelling study predicts

Research suggests that the combination of incentives to publish and the use of large language models will lead to more papers, but they will be less refined.

nature.com

Another nail in the coffin of fetishisation of large data. I applaud the careful work that goes into this, but averaging physiological detail that in fact fluctuates at multiple timescales, all relevant to thought & behaviour, was never yielding insight. www.thetransmitter.org/sex-differen...

Brain’s sex differences are subtle and contradictory, large MRI study finds

Sex-based behavioral differences do not match with variations in brain activation or structure in a study of almost 1,000 people.

thetransmitter.org

Wilson seems to review a dismal literary adaptation rather than a blockbuster epic in @lrb.co.uk. I agree Nolan's dialogue-driven character development is poor, but she misreads the Trojan Horse sequence & Sinon's willing sacrifice - a smart cinematic device in my view. www.lrb.co.uk/the-paper/v4...

Emily Wilson · An Uncomplicated Man

As they say at the awards shows, I was humbled to learn that Christopher Nolan has read at least the first line of my...

lrb.co.uk

With respect, editors must consider whether their 'expert' on the future of science is in fact a practising scientist. Trawling data for regression coefficients, irrespective of the field of application, is applied computer science, i.e. engineering. Science and engineering are different things.