Reposted by Fernando Pérez-GarcíaMoira Donegan @moiradonegan.bsky.social · 23/02/2026Why do I have to pretend that I'm going to print something in order to save it as a PDF. Why do I have to engage in a little ruse. 334191442875
Fernando Pérez-García @fepegar.com · 30/01/2026We are excited to release the weights of @msftresearch.bsky.social's COLIPRI, our 3D vision–language encoder for chest CT scans, on @hf.co 🤗 Model: aka.ms/colipri Demo: aka.ms/colipri-demo Paper: aka.ms/colipri-paper Why does COLIPRI matter? 🧵0/12 👇 132
Reposted by Fernando Pérez-GarcíaAndrew Heiss @andrew.heiss.phd · 15/12/2025Grading and googling hallucinated citations, as one does nowadays, and now that LLMs have been around for a while, I've discovered new horrors: hallucinated journals are now appearing in Google Scholar with dozens of citations bc so many people are citing these fake things 13039551264
Reposted by Fernando Pérez-GarcíaGautam Kamath @gautamkamath.com · 07/12/2025This article frames the problem as [AI?] slop. Which is a problem, but not the main one here. This is an issue with authorship norms and practices. A single individual putting their name on hundreds of (workshop) papers they admitted they had little part in. 3213
Reposted by Fernando Pérez-GarcíaDaniel Coelho de Castro @dccastr0.bsky.social · 07/12/2025Excited to speak shortly at the Medical Imaging at @euripsconf.bsky.social workshop in Copenhagen, where I'll share some insights from our @msftresearch.bsky.social team's journey "From medical image interpretation to scientific discovery" over the past couple of years. 022
Reposted by Fernando Pérez-GarcíaLukas C. H. @gewoonlukas.bsky.social · 26/11/2025SpaceX is ready for its next Transporter mission! With 140 satellites onboard, this is the largest Transporter mission since Transporter-1 in 2021, which carried 143 satellites. Here's my identification attempt. Launch is scheduled for NET 18:19 UTC. 269737
Reposted by Fernando Pérez-GarcíaPython Software Foundation @python.org · 27/10/2025TLDR; The PSF has made the decision to put our community and our shared diversity, equity, and inclusion values ahead of seeking $1.5M in new revenue. Please read and share. pyfound.blogspot.com/2025/10/NSF-... 🧵python.orgThe official home of the Python Programming Language 12363772727
Reposted by Fernando Pérez-GarcíaMax Ilse @maxilse.bsky.social · 23/09/2025🩻Excited to share our latest preprint: “Data Scaling Laws for Radiology Foundation Models” Foundation vision encoders like CLIP and DINOv2 have transformed general computer vision, but what happens when we scale them for medical imaging? 📄 Read the full preprint here: arxiv.org/abs/2509.12818arxiv.orgData Scaling Laws for Radiology Foundation ModelsFoundation vision encoders such as CLIP and DINOv2, trained on web-scale data, exhibit strong transfer performance across tasks and datasets. However, medical imaging foundation models remain constrai... 152
Reposted by Fernando Pérez-GarcíaAmnesty International UK @amnestyuk.bsky.social · 22/09/2025Recognition is no doubt significant but it will be a hollow gesture if the UK does not also seek to end Israel's genocide, illegal occupation, and system of apartheid against the Palestinian people. 🧵1/3 511560
Reposted by Fernando Pérez-GarcíaChristopher Owen @oysta.au · 21/09/2025Yeah man we should really fight back by staying on X 256169914055
Fernando Pérez-García @fepegar.com · 21/08/2025Having some fun with DINOv3 and PCA! Although I'm not happy my nose has such a low foreground probability :D 060
Reposted by Fernando Pérez-GarcíaSam Rose @samwho.dev · 17/08/2025Email addresses are very simple, and you will score highly in this quiz. e-mail.wtfe-mail.wtfEmail is EasyEveryone knows what an email address is, right? 38274128
Reposted by Fernando Pérez-GarcíaMax Seitzer @maxseitzer.bsky.social · 14/08/2025Introducing DINOv3 🦕🦕🦕 A SotA-enabling vision foundation model, trained with pure self-supervised learning (SSL) at scale. High quality dense features, combining unprecedented semantic and geometric scene understanding. Three reasons why this matters👇 2269
Reposted by Fernando Pérez-GarcíaMerriam-Webster @merriam-webster.com · 08/08/2025We’re not sure who needs to hear this, but ‘blueberry’ has two b’s. 2337128918
Reposted by Fernando Pérez-GarcíaOlivia Guest · Ολίβια Γκεστ @olivia.science · 29/07/2025Boiling here at home in Cyprus but I put the finishing touches a couple of days ago on this preprint: What Does 'Human-Centred AI' Mean? doi.org/10.48550/arX... Wherein I analyse HCAI & demonstrate through 3 triplets my new tripartite definition of AI (Table 1) that properly centres the human. 1/n 11419122
Reposted by Fernando Pérez-GarcíaNature @nature.com · 23/07/2025Scientists overwhelmingly recognize the value of sharing null results, but rarely publish them in the research literature go.nature.com/450KElrgo.nature.comResearchers value null results, but struggle to publish themSurvey finds that fear of reputational harm and a lack of support and publication platforms are among respondents’ key concerns. 513766
Reposted by Fernando Pérez-GarcíaEllie Huxtable @ellie.wtf · 23/07/2025i love it when people make PRs really easy to review ❤️ feels really bad when there's a PR that's been open for a while, but i know i can't easily do a good job of reviewing it, so it's constantly "later" 2221
Reposted by Fernando Pérez-GarcíaDrBreaky @drbreaky.bsky.social · 21/07/2025New 2 year position as an Imaging Data Infrastructure Specialist Located a the Hunter Medical Research Institute (HMRI) & the University of Newcastle (UoN), in partnership with the National Imaging Facility (NIF), Apply here: www.seek.com.au/job/85537666 0178
Reposted by Fernando Pérez-GarcíaKosta Derpanis @csprofkgd.bsky.social · 19/07/2025A #computervision researcher at #ICML2025 3657
Reposted by Fernando Pérez-GarcíaStephanie Hyland @hylandsl.bsky.social · 18/07/2025New work from my team! arxiv.org/abs/2507.12950 Intersecting mechanistic interpretability and health AI 😎 We trained and interpreted sparse autoencoders on MAIRA-2, our radiology MLLM. We found a range of human-interpretable radiology reporting concepts, but also many uninterpretable SAE features.arxiv.orgInsights into a radiology-specialised multimodal large language model with sparse autoencodersInterpretability can improve the safety, transparency and trust of AI models, which is especially important in healthcare applications where decisions often carry significant consequences. Mechanistic... 1114
Reposted by Fernando Pérez-GarcíaJournal of Open Source Software @joss-openjournals.bsky.social · 11/07/2025Just published in JOSS: 'Lighter: Configuration-Driven Deep Learning' doi.org/10.21105/joss.08101 021
Reposted by Fernando Pérez-GarcíaMicrosoft Research @msftresearch.bsky.social · 26/06/2025The world’s first multimodal, bilingual radiology dataset could reshape the way radiologists and AI systems make sense of X-rays. PadChest-GR, developed by the University of Alicante with Microsoft Research, has the potential to advance research across the field for years to come. msft.it/6013SLDYZ 073
Reposted by Fernando Pérez-GarcíaDaniel Coelho de Castro @dccastr0.bsky.social · 26/06/2025📣 Our new PadChest-GR benchmark in collaboration with @ua.es and MedBravo is published in @ai.nejm.org: ai.nejm.org/doi/full/10....! 📰 Also check out our blog post for why we're so excited about it: www.microsoft.com/en-us/resear... 💾 The dataset can be downloaded here: bimcv.cipf.es/bimcv-projec...ai.nejm.orgPadChest-GR: A Bilingual Chest X-Ray Dataset for Grounded Radiology Report GenerationArtificial intelligence (AI)–powered radiology report generation (RRG) aims to create free-text radiology reports from clinical imaging. Grounded radiology report generation (GRRG) augments RRG by ... 062
Reposted by Fernando Pérez-GarcíaNEJM AI @ai.nejm.org · 03/07/2025A novel dataset designed to train and evaluate grounded report generation models from chest x-ray images includes comprehensive sentence-level bounding-box annotations for all clinically relevant findings in each image. Learn more: nejm.ai/44rk1a1 @dccastr0.bsky.social #AI #MedSky #MLSky 044
Reposted by Fernando Pérez-Garcíapr0grammerhum0r.bsky.social @pr0grammerhum0r.bsky.social · 27/06/2025codingAssistantsAreJustCasinosForProgrammers 081
Fernando Pérez-García @fepegar.com · 26/06/2025Congratulations/enhorabuena to @dccastr0.bsky.social and the rest of the team 👏 One important step towards usability of AI in real-world radiology applications. 060
Reposted by Fernando Pérez-Garcíaconda-forge @conda-forge.org · 12/06/2025Demystifying conda packages! 021
Reposted by Fernando Pérez-GarcíaKosta Derpanis @csprofkgd.bsky.social · 18/05/2025A new use of the asterisk in the paper author list for credit assignment 841266
Reposted by Fernando Pérez-GarcíaFrançois Fleuret @francois.fleuret.org · 28/04/2025I asked "on the other platform" what were the most important improvements to the original 2017 transformer. That was quite popular and here is a synthesis of the responses: 420643
Reposted by Fernando Pérez-GarcíaEugene Vinitsky 🍒 @eugenevinitsky.bsky.social · 26/04/2025Some of the anti-AI stuff feels a bit like when people would say "don't use Wikipedia as a source." It's just like anything else, a piece of information that you weigh against multiple sources and your own understanding of its likely failure modes 55444838
Reposted by Fernando Pérez-GarcíaSander Dieleman @sedielem.bsky.social · 25/04/2025One weird trick for better diffusion models: concatenate some DINOv2 features to your latent channels! Combining latents with PCA components extracted from DINOv2 features yields faster training and better samples. Also enables a new guidance strategy. Simple and effective! 0284
Reposted by Fernando Pérez-GarcíaMauro Entrialgo @tyrexito.bsky.social · 21/04/2025Al papa lo que es del papa. 462748887
Reposted by Fernando Pérez-GarcíaFélix López Luis 🇺🇦 @flopezluis.bsky.social · 16/04/2025I wrote this to help my team build an Operational Excellence mindset. I actually shared it internally with the whole company. Sharing it here in case it's useful to someone: docs.google.com/document/d/1...docs.google.comOperational ExcellenceOperational Excellence Operational excellence is a mindset and set of practices focused on delivering highly reliable systems with continuous improvement. It blends engineering quality, observability,... 42612
Reposted by Fernando Pérez-GarcíaSimon Willison @simonwillison.net · 18/04/2025Gemini 2.5 Pro and Flash now have the ability to return image segmentation masks on command, as base64 encoded PNGs embedded in JSON strings I vibe coded an interactive tool for exploring this new capability - it costs a fraction of a cent per image simonwillison.net/2025/Apr/18/... 71437
Reposted by Fernando Pérez-GarcíaDavid Gasquez @davidgasquez.com · 14/03/2025A while back, I worked on some tooling to export all Spain's National Institute of Statistics datasets to Parquet and published them to @hf.co. davidgasquez.com/exporting-in... I'm now looking for ways to run this weekly, ideally, on free infrastructure. Ideas?davidgasquez.comWith the information above, I started to tinker with different methods to export as many tables as possible and settled on the following approach. 4132
Reposted by Fernando Pérez-GarcíaInternet Archive @archive.org · 13/03/2025🔗 Websites vanish. Links break. But knowledge can live on with your help. Use the Wayback Machine’s Save Page Now tool to archive webpages that are important to you. 🕰️ 💾 📌 Try it now: web.archive.org/save 6148872062
Fernando Pérez-García @fepegar.com · 01/03/2025Inspired by itry, I’ve written jtry. It creates a temporary folder with a virtual environment managed by uv containing the passed packages and an empty Jupyter notebook, then opens the notebook in VS Code. Try e.g. curl raw.githubusercontent.com/fepegar/jtry... | bash -s -- transformers pydanticraw.githubusercontent.com 040
Reposted by Fernando Pérez-GarcíaSimon Willison @simonwillison.net · 14/02/2025Quick TIL: if you have uv installed and you want to try out ANY Python package in an ipython REPL, you can do so just by running uvx --with llm --with sqlite-utils ipython Pass as many --with package options as you like I wrapped it in a shell script, "itry llm" til.simonwillison.net/python/itrytil.simonwillison.netTrying out Python packages with ipython and uvxI figured out a really simple pattern for experimenting with new Python packages today: 515820
Reposted by Fernando Pérez-GarcíaHumanists UK @humanists.uk · 23/01/2025These are the countries that have already banned hitting children. We're supporting a push from @jessasatomp.bsky.social to add England to this list. 1329
Reposted by Fernando Pérez-GarcíaNature @nature.com · 23/01/2025CERN will provide financial incentives to encourage practices such as data sharing and transparent peer review go.nature.com/4geENfMgo.nature.comPioneering CERN scheme will pay publishers more if they hit open-science targetsThe particle-physics funder will provide financial incentives to encourage practices such as data sharing and transparent peer review. 07814
Reposted by Fernando Pérez-GarcíaHannibal Lecter 🦋 @hannibal.bsky.social · 21/01/2025Para que quede claro. 15130591663
Reposted by Fernando Pérez-GarcíaOtto Mas 🎙️ @ottomas.es · 21/01/2025Pues para no ser un pato, lo parece 612408963
Fernando Pérez-García @fepegar.com · 24/12/2024Great read. “Once there is a concrete path to revenue […], it becomes increasingly more difficult for researchers to continue to ask for full research freedom. Many will be asked to directly contribute to products […], and only a few will be allowed continued freedom in research.” 020
Reposted by Fernando Pérez-GarcíaEthan Mollick @emollick.bsky.social · 16/12/2024I gave most of the frontier models this prompt: "create something I can paste into p5js that will startle me with its cleverness in creating something that invokes the control panel of a starship in the distant future" Here is what they did (I didn't include Gemini 1.5, it kept making errors) 711410
Reposted by Fernando Pérez-GarcíaStella Li @stellali.bsky.social · 06/12/202431% of US adults use generative AI for healthcare 🤯But most AI systems answer questions assertively—even when they don’t have the necessary context. Introducing #MediQ a framework that enables LLMs to recognize uncertainty🤔and ask the right questions❓when info is missing: 🧵 36814
Reposted by Fernando Pérez-GarcíaStephanie Hyland @hylandsl.bsky.social · 12/12/2024Lidong Zhou giving lessons on scaling from the systems community #NeurIPS2024 061