iiinova — where your digital self keeps evolving

Lorenzo Bertolini @lorenzoscottb.bsky.social · 22/05/2026

Hello #sleeppeeps, our latest work on AI for dream report analysis is finally out (with attached pre-trained models). In the work, we show how small generative LLMs can be easily trained to efficiently annotate dream reports in a more dynamic manner. link.springer.com/chapter/10.1...

link.springer.com

PreDA: Prefix-Based Dream Reports Annotation with Generative Language Models

Dream reports are recollections of our experiences while asleep, and have strong research and clinical value. Since their analysis can be extremely time-consuming, researchers have adopted multiple ty...

021

Reposted by Lorenzo Bertolini

Adriana Michalak, PhD @dr-adri.bsky.social · 24/03/2026

💤𝗛𝗼𝘁 𝗼𝗳𝗳 𝘁𝗵𝗲 𝗽𝗿𝗲𝘀𝘀💤 Our findings indicate that 𝘀𝘂𝗯𝗷𝗲𝗰𝘁𝗶𝘃𝗲 𝘀𝗹𝗲𝗲𝗽 𝗱𝗲𝗽𝘁𝗵 is not solely determined by reduced brain activity. 𝗜𝗺𝗺𝗲𝗿𝘀𝗶𝘃𝗲 𝗱𝗿𝗲𝗮𝗺 𝗲𝘅𝗽𝗲𝗿𝗶𝗲𝗻𝗰𝗲𝘀 appear to maintain the 𝘀𝘂𝗯𝗷𝗲𝗰𝘁𝗶𝘃𝗲 𝘀𝗲𝗻𝘀𝗲 𝗼𝗳 𝗱𝗲𝗲𝗽 𝘀𝗹𝗲𝗲𝗽 as 𝗵𝗼𝗺𝗲𝗼𝘀𝘁𝗮𝘁𝗶𝗰 𝘀𝗹𝗲𝗲𝗽 𝗽𝗿𝗲𝘀𝘀𝘂𝗿𝗲 naturally declines across the night. dx.plos.org/10.1371/jour...

dx.plos.org

Immersive NREM2 dreaming preserves subjective sleep depth against declining sleep pressure

Perceived sleep depth is generally thought to reflect reduced brain activity. This study shows that this relationship weakens during dreaming, suggesting that dreaming helps sustain the subjective exp...

161

Lorenzo Bertolini @lorenzoscottb.bsky.social · 08/01/2026

👋 #sleeppeeps, this weekend I’ll be at the DxE simposyum at MIT to present some preliminary work at the intersection of AI and neuroscience to trace #dream event in event #EEG signal. Come say hi at the event, or drop a line if you’re in the Boston area and wanna chat 😀 dxe.framer.website

dxe.framer.website

Dream x Engineering Symposium

A gathering of world leaders in sleep and dream science exploring how we can influence dreaming with precision to improve well-being, accelerate learning, and treat clinical conditions.

030

Lorenzo Bertolini @lorenzoscottb.bsky.social · 17/11/2025

All the models are back online, and the full paper should be out soon too 😃 huggingface.co/collections/... #sleeppees #dream

huggingface.co

PreDA - a DReAMy-lib Collection

generative models for multi-feature annotation

031

Lorenzo Bertolini @lorenzoscottb.bsky.social · 03/08/2025

Unserious post: Now that (some) #AI researchers and engineers have rockstars-level salaries and notoriety, will they also have stage (or, better, paper) names?

021

Lorenzo Bertolini @lorenzoscottb.bsky.social · 01/08/2025

Made it alive and well to the last day of #ACL2025NLP? Hats off you to you (well…us…)!!! Time to collect the last energy and pass by our poster this afternoon 😀

010

Lorenzo Bertolini @lorenzoscottb.bsky.social · 27/07/2025

Not pressing _this_ work at #ACL2025NLP, but always happy to chat NLP/AI for dream analysis 😴🤖

020

Lorenzo Bertolini @lorenzoscottb.bsky.social · 25/07/2025

I’ll be at #ACL2025NLP next week, with a paper for the #GeBNLP workshop! In this work, we tested the ability of a task-level explainability method to trace biological-sex biases in medical image classification, using general and biomedical VLMs. @aclmeeting.bsky.social @genderbiasnlp.bsky.social

130

Lorenzo Bertolini @lorenzoscottb.bsky.social · 23/07/2025

👋 #sleeppeeps In this new work, we show how multiple LLMs appear to easily process #dreams, and reflect previously observed differences in reports (e.g., male-female, blind-sighted). We also argue for caution linking dreams and bizarreness when using #LLMs www.frontiersin.org/journals/sle...

frontiersin.org

Frontiers | Dreams are more “predictable” than you think

IntroductionA growing body of work has used machine learning and AI tools to analyse dream reports, and compare them to other textual content. Since these to...

071

Lorenzo Bertolini @lorenzoscottb.bsky.social · 27/06/2025

Happy to share that our work on #conformalprediction for pulmonary nodule detection has been accepted for presentation at COPA 25 conference!

000

Lorenzo Bertolini @lorenzoscottb.bsky.social · 15/05/2025

Introducing PreDA (prefix-based dream annotation), a set of generative LLMs tuned to annotate dream reports for 6 Hall & Van De Castle features! The paper will be presented at LOD 25 conference later this year as a long contribution. Models available here huggingface.co/jrc-ai #sleeppeeps #dream

021

Reposted by Lorenzo Bertolini

arxiv cs.CV @arxiv-cs-cv.bsky.social · 31/12/2024

Roel Hulsman, Valentin Comte, Lorenzo Bertolini, Tobias Wiesenthal, Antonio Puertas Gallardo, Mario Ceresa Conformal Risk Control for Pulmonary Nodule Detection arxiv.org/abs/2412.20167

051

Reposted by Lorenzo Bertolini

Adriana Michalak, PhD @dr-adri.bsky.social · 18/12/2024

💤 As I approach the end of my PostDoc at IMT Lucca, I’m looking for a new job! 🧠 Ideally: Research roles related to sleep/dream/consciousness/cognitive science. 🤖 Open to scientific writing & data analysis, too. 📍 Milan, nearby areas, or Switzerland. DMs open 💃🏼

0106

Reposted by Lorenzo Bertolini

Tomer Ullman @tomerullman.bsky.social · 01/12/2024

thinking of calling this "The Illusion Illusion" (more examples below)

601570381

Reposted by Lorenzo Bertolini

François Fleuret @francois.fleuret.org · 26/11/2024

My deep learning course at the University of Geneva is available on-line. 1000+ slides, ~20h of screen-casts. Full of examples in PyTorch. fleuret.org/dlc/ And my "Little Book of Deep Learning" is available as a phone-formatted pdf (nearing 700k downloads!) fleuret.org/lbdl/

461253247

Reposted by Lorenzo Bertolini

Thomas Wolf @thomwolf.bsky.social · 24/11/2024

It's Sunday morning so taking a minute for a nerdy thread (on math, tokenizers and LLMs) of the work of our intern Garreth By adding a few lines of code to the base Llama 3 tokenizer, he got a free boost in arithmetic performance 😮 [thread]

527134

Reposted by Lorenzo Bertolini

Trish Greenhalgh @trishgreenhalgh.bsky.social · 24/11/2024

Reproducing my thread from the Other Place on our Lancet paper 'Long COVID - a clinical update'. Coauthors Manoj Sivan, Alice Perlowski, Janko Nikolich. 1/ www.thelancet.com/journals/lan...?

thelancet.com

Long COVID: a clinical update

Post-COVID-19 condition (also known as long COVID) is generally defined as symptoms persisting for 3 months or more after acute COVID-19. Long COVID can affect multiple organ systems and lead to sever...

831313557

Reposted by Lorenzo Bertolini

il Post @ilpost.it · 21/11/2024

E quindi come si sta da queste parti?

22669252

Reposted by Lorenzo Bertolini

AmsterdamNLP @amsterdamnlp.bsky.social · 10/11/2024

Work in progress -- suggestions for NLP-ers based in the EU/Europe & already on Bluesky very welcome! go.bsky.app/NZDc31B

517220

Lorenzo Bertolini @lorenzoscottb.bsky.social · 20/11/2024

Hi alI quick dataset question: I’m looking for a collection of histopathological images (no specific requirements wrt organ/disease/task) which includes metadata on participants’ (approximate) age. Any suggestion?

140

Reposted by Lorenzo Bertolini

Manlio De Domenico @manlius.bsky.social · 15/11/2024

As a computational social scientist I could not miss to follow this beautiful creature: the BlueSky social network during its growth in the last hours! In this video I show just a snapshot for about 210k users and their 500k-ish connections #academicsky Zoom in the next post youtu.be/JcmYZtWGnjQ

youtu.be

The “great Xodus” and the growth of BlueSky

YouTube video by Manlio De Domenico

67032

Lorenzo Bertolini @lorenzoscottb.bsky.social · 12/11/2024

Nice to see so many new #sleeppeeps on #SleepSky! I’m just gonna leave this here again 😅 (btw the paper and DReAMy abstract have now been published!)

041

Reposted by Lorenzo Bertolini

Christopher Manning @chrmanning.bsky.social · 10/11/2024

Papers at #EMNLP2024 #1 Statistical Uncertainty in Word Embeddings: GloVe-V Neural models, from word vectors through transformers, use point estimate representations. They can have large variances, which often loom large in CSS applications. Tue Nov 12 15:15-15:30 Flagler

05713

Reposted by Lorenzo Bertolini

Maike Osborne @maosbot.bsky.social · 09/11/2024

New here? Interested in AI/ML? Check out these great starter packs! AI: go.bsky.app/SipA7it RL: go.bsky.app/3WPHcHg Women in AI: go.bsky.app/LaGDpqg NLP: go.bsky.app/SngwGeS AI and news: go.bsky.app/5sFqVNS You can also search all starter packs here: blueskydirectory.com/starter-pack...

66557212

Reposted by Lorenzo Bertolini

Juan Diego Rodriguez @juand-r.bsky.social · 08/11/2024

How do language models organize concepts and their properties? Do they use taxonomies to infer new properties, or infer based on concept similarities? Apparently, both! 🌟 New paper with my fantastic collaborators @amuuueller.bsky.social and @kanishka.bsky.social

Title: "Characterizing the Role of Similarity in the Property Inferences of Language Models"
Authors: Juan Diego Rodriguez, Aaron Mueller, Kanishka Misra

Left figure: "Given that dogs are daxable, is it true that corgis are daxable?" A language model could answer this either using taxonomic relations, illustrated by a taxonomy dog-corgi, dog-mutt, canine-wolf, etc., or by similarity relations (dogs are more similar to corgis than cats, wolves or shar peis).

Right figure: illustration of the causal model (and an example intervention) for distributed alignment search (DAS), which we used to find a subspace in the network responsible for property inheritance behavior. The bottom nodes are "property", "premise concept (A)" and "conclusion concept (B)" , the middle nodes are "A has property P", "B is a kind of A", and the top node is "B has property P".

410822

Lorenzo Bertolini @lorenzoscottb.bsky.social · 05/06/2024

Sad to miss the #neuroscience of #dreaming summers school and the opportunity to chat about #NLP and #LLMs for #dream research. If you're interested, check out dreamy's tutorials on the Git page (also available on Google Colab) 😀 github.com/lorenzoscott...

github.com

GitHub - lorenzoscottb/DReAMy: DReAMy: a library for dream-reports annotation methods with python, NLP, and LLMs

DReAMy: a library for dream-reports annotation methods with python, NLP, and LLMs - lorenzoscottb/DReAMy

031

Lorenzo Bertolini @lorenzoscottb.bsky.social · 23/05/2024

👋 #SleepPeeps, happy to share DReAMy's new feature! Do you need to anonymise #dream reports? We've got you covered, with anonymized, a 1-line solution to find and replace entities. Try it out in our tutorial section, or online colab.research.google.com/drive/14hHRRC3…

010

Lorenzo Bertolini @lorenzoscottb.bsky.social · 11/04/2024

Hello friends and #sleeppeeps, later this month I will be giving an talk on my (and relevant) work on #NLP, #LLMs and #dreams. Catch it of you're interested

010

Lorenzo Bertolini @lorenzoscottb.bsky.social · 31/01/2024

Hello #sleeppeeps, happy to share that our paper "Automatic Annotation of #Dream Report’s Emotional Content with Large Language Models" was accepted for publication at #EACL2024 Computational Linguistic and Clinical Psychology Workshop! Here's a short 🧵

100

Reposted by Lorenzo Bertolini

Carl T. Bergstrom @carlbergstrom.com · 14/12/2023

There's a curious new rhetorical strategy I'm encountering more and more often in online spaces wherein ignorance about a claim is taken to refute it. It goes a bit like the script below. It's the death-of-expertise script, but with an additional twist. Has anyone written about this?

Approach A has problem B.

You’re wrong and I’m not going to fall for that.

Why do you say that?

Because I don’t know enough about A or B to draw that conclusion, and I’m an Open-Minded Thinker.

Oh, so you’re not saying that you disagree, you’re saying that you don’t know.

No, you’re wrong. You say A has problem B — but without knowing about A and B, it’s obvious I can’t conclude that.

So A and B is what I happen to work on.

Now you are bullying me with credentialism.

Not at all. I just wanted to suggest some background that will explain why A has problem B.

I know better than to read that background.

What? Why?

You’re the one claiming A has problem B. So I’m not going to trust your evidence for anything. I’m not One Of Those Easy Dupes.

2219641

Lorenzo Bertolini @lorenzoscottb.bsky.social · 13/11/2023

Mondays are so much better when you wake up to interesting experimental results!

000

Lorenzo Bertolini @lorenzoscottb.bsky.social · 24/10/2023

Quite amazing to see people are already using DReAMy in the their research at #WorldSleep2023 !!! Don't forget to stop by DReAMy's oral presentation (Room A07, 4:58) if you want to learn more about #LLMs for #dream reports analysis and annotation!

020

Lorenzo Bertolini @lorenzoscottb.bsky.social · 24/10/2023

Hey #SleepPeeps, this is your friendly reminder to pass by our presentation later this afternoon, if you're at #WorldSleep2023

010

Lorenzo Bertolini @lorenzoscottb.bsky.social · 19/10/2023

Hello #sleeppeeps! #WorldSleep2023 is a day away. Don't forget to stop by our oral presentation. We'll talk about our experiments with #LLMs on #dream reports, the library itself, and how to use it. Don't #sleep on it 😀

010

Lorenzo Bertolini @lorenzoscottb.bsky.social · 12/10/2023

On a non-so-sciency note, we're planning on handing out some DReAMy stickers at #WorldSleep23 ! #scisky #sleeppeeps

020

Lorenzo Bertolini @lorenzoscottb.bsky.social · 06/10/2023

Hello, #SleepSky, #SleepPeeps, and #dream researchers. It's nice to see many of you here, and reciprocating the interest of a curious #NLP/#AI researcher. So I thought to do a quick thread about my interested dreams and sleep, and what I've done so far!

141

Lorenzo Bertolini @lorenzoscottb.bsky.social · 04/10/2023

Hi #sleeppeeps, I will be around #eSLEEPEurope23 virtual congress this week. Shout out if you want to chat about #NLP and #LLMs for #dream analysis during one of the breaks 😀

000

Lorenzo Bertolini @lorenzoscottb.bsky.social · 03/10/2023

Hello #sleeppeeps and #SciSky! Later this week, DReAMy and I will be at #eSleepEurope23, with a poster on how to train generative #LLMs to annotate #dream reports for different Hall & Van de Castle features. You can find models and demo at huggingface.co/DReAMy-lib

010

Lorenzo Bertolini @lorenzoscottb.bsky.social · 02/10/2023

#HiSciSky, Lorenzo here! I am researcher at the European Commission's JRC, working on biomedical NLP. Before, I did a PhD and postdoc at Sussex Uni, where I focused on topics like compositionality, graph #ML, and dream analysis.

050