Robbert Struyven

11 posts

Robbert Struyven

Robbert Struyven

@RobbertStruyven

Medical Imaging PhD student at @UCL & @Moorfields Eye Hospital

London, UK Katılım Kasım 2018
580 Takip Edilen148 Takipçiler
Robbert Struyven retweetledi
Rishi Sunak
Rishi Sunak@RishiSunak·
We're already using AI to get people the care they need quicker. Last week at @Moorfields I saw how it can predict life-threatening conditions like strokes. Yesterday @SteveBarclay announced the 64 NHS trusts receiving funding to boost the use of AI. twitter.com/SteveBarclay/s…
Steve Barclay@SteveBarclay

We’re rolling out more cutting-edge AI tech across the NHS for quicker, more accurate diagnosis of #LungCancer. 64 NHS trusts will benefit from £21M to do this. Speaking with trusts and healthcare leaders today, we discussed the safe deployment of AI to help cut waiting lists.

English
26
13
59
15.5K
Robbert Struyven retweetledi
Matthieu Meeus
Matthieu Meeus@matthieu_meeus·
Excited about our most recent work! We propose a method to infer whether an LLM has seen a specific document during training or not. See 🧵 for details.
Yves-A. de Montjoye@yvesalexandre

🤔 Did an LLM read your book 📘? In a new @arxiv preprint, we propose the document-level membership inference task for LLMs. Using only the LLM output, we can accurately tell if a document was seen during training (AUC=0.856) arxiv.org/abs/2310.15007. A short 🧵

English
0
1
3
256
Robbert Struyven
Robbert Struyven@RobbertStruyven·
@yvesalexandre @arxiv @tegmark new preprint enabling you to actually test if your books were used without your permission during LLM training twitter.com/tegmark/status…
Max Tegmark@tegmark

Dear Margaret @mmitchell_ai: You're the Chief Ethics Scientist of @huggingface, which without my permission republishes my copyrighted books for Big Tech AI training that I oppose. This is unethical and illegal – can you please help me resolve this?

English
0
0
2
163
Yves-A. de Montjoye
Yves-A. de Montjoye@yvesalexandre·
🤔 Did an LLM read your book 📘? In a new @arxiv preprint, we propose the document-level membership inference task for LLMs. Using only the LLM output, we can accurately tell if a document was seen during training (AUC=0.856) arxiv.org/abs/2310.15007. A short 🧵
English
4
12
36
9K
Robbert Struyven retweetledi
Rishi Sunak
Rishi Sunak@RishiSunak·
AI could solve problems we once thought beyond us. Today @Moorfields, I saw how they’re using AI to predict heart attacks and strokes. We’re investing a further £100m to accelerate the use of AI in healthcare, supporting breakthroughs for previously incurable diseases.
Rishi Sunak tweet media
English
1.1K
133
599
183.9K
Robbert Struyven retweetledi
Nathan Benaich
Nathan Benaich@nathanbenaich·
🪩The @stateofai 2023 is now here. Our 6th installment is one of the most exciting years I can remember. The #stateofai report covers everything you *need* to know, covering research, industry, safety and politics. There’s lots in there, so here’s my director’s cut 🧵
Nathan Benaich tweet media
English
57
489
1.6K
970.9K
Robbert Struyven retweetledi
Demis Hassabis
Demis Hassabis@demishassabis·
Huge congrats to @pearsekeane & teams at @INSIGHTeyehub @Moorfields @UCLeye! This freely available AI model can help identify sight-threatening diseases & other issues from heart attacks to Parkinson’s disease. Thrilled to see continued impact of our longstanding collaboration.
INSIGHTeyehub@INSIGHTeyehub

A world-first #AI foundation model for eyes, just out in @Nature, is set to catalyse research into preventing blindness. The work from @Moorfields @UCLeye and enabled by INSIGHT builds on a 2016 collab with @GoogleDeepMind @pearsekeane @DemisHassabis bit.ly/3rbqup0

English
11
39
199
51.5K
Robbert Struyven retweetledi
Yukun Zhou
Yukun Zhou@Yukunzhou19·
We are thrilled to share that our new retinal foundation model, "RETFound" published today in @Nature! To our knowledge, RETFound is the first #opensource foundation model for retinal images and one of the earliest across all medical imaging fields. nature.com/articles/s4158…
Eric Topol@EricTopol

Just published @Nature 1st major large language model in medicine Self-supervised #AI of 2 million retina images to predict many diseases beyond the eye, such as heart attack, stroke, heart failure, Parkinson's disease nature.com/articles/s4158… led by @pearsekeane @Yukunzhou19

English
5
27
105
15.3K