Max Ilse
@maxilse.bsky.social
Working at microsoft research health futures. Interested in causal representation learning and generative modelling applied to medical data.
🩻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/2509.12818
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/2509.12818
Data Scaling Laws for Radiology Foundation Models
Foundation 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...
arxiv.org
September 23, 2025 at 8:34 AM
🩻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/2509.12818
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/2509.12818
What a damning abstract
Some interesting detective work alleging that proprietary LLM developers are gaming the Chatbot Arena leaderboards, with collusion from Chatbot Arena's operators.
The Leaderboard Illusion
Measuring progress is fundamental to the advancement of any scientific field. As benchmarks play an increasingly central role, they also grow more susceptible to distortion. Chatbot Arena has emerged ...
arxiv.org
April 30, 2025 at 8:28 AM
What a damning abstract
Reposted by Max Ilse
I want to reshare @brandfonbrener.bsky.social's @NeurIPSConf 2024 paper on CoLoR-Filter: A simple yet powerful method for selecting high-quality data for language model pre-training!
With @hlzhang109.bsky.social @schwarzjn.bsky.social @shamkakade.bsky.social
With @hlzhang109.bsky.social @schwarzjn.bsky.social @shamkakade.bsky.social
April 5, 2025 at 12:04 PM
I want to reshare @brandfonbrener.bsky.social's @NeurIPSConf 2024 paper on CoLoR-Filter: A simple yet powerful method for selecting high-quality data for language model pre-training!
With @hlzhang109.bsky.social @schwarzjn.bsky.social @shamkakade.bsky.social
With @hlzhang109.bsky.social @schwarzjn.bsky.social @shamkakade.bsky.social
Reposted by Max Ilse
⚫⚪ It's coming...SHADES. ⚪⚫
The first ever resource of multilingual, multicultural, and multigeographical stereotypes, built to support nuanced LLM evaluation and bias mitigation. We have been working on this around the world for almost **4 years** and I am thrilled to share it with you all soon.
The first ever resource of multilingual, multicultural, and multigeographical stereotypes, built to support nuanced LLM evaluation and bias mitigation. We have been working on this around the world for almost **4 years** and I am thrilled to share it with you all soon.
February 10, 2025 at 8:28 AM
⚫⚪ It's coming...SHADES. ⚪⚫
The first ever resource of multilingual, multicultural, and multigeographical stereotypes, built to support nuanced LLM evaluation and bias mitigation. We have been working on this around the world for almost **4 years** and I am thrilled to share it with you all soon.
The first ever resource of multilingual, multicultural, and multigeographical stereotypes, built to support nuanced LLM evaluation and bias mitigation. We have been working on this around the world for almost **4 years** and I am thrilled to share it with you all soon.
Reposted by Max Ilse
We’re looking for a motivated researcher to apply for a Marie Skłodowska-Curie postdoc with our Econometrics & Data Science group at SDU!
Focus: Causal Inference, Machine Learning, Big Data
Full support for promising projects
More info & apply:
www.sdu.dk/en/om-sdu/in...
Focus: Causal Inference, Machine Learning, Big Data
Full support for promising projects
More info & apply:
www.sdu.dk/en/om-sdu/in...
Econometrics and Data Science
www.sdu.dk
January 30, 2025 at 7:43 AM
We’re looking for a motivated researcher to apply for a Marie Skłodowska-Curie postdoc with our Econometrics & Data Science group at SDU!
Focus: Causal Inference, Machine Learning, Big Data
Full support for promising projects
More info & apply:
www.sdu.dk/en/om-sdu/in...
Focus: Causal Inference, Machine Learning, Big Data
Full support for promising projects
More info & apply:
www.sdu.dk/en/om-sdu/in...
Reposted by Max Ilse
Apply!
Assistant Professor (L/SL) in AI, including computer vision [DL 5 Mar] @BristolUni - awarded AI University of the Year in 2024.
DM to myself or @_SethBullock_ for inquiries (don't us send CV pls!, apply directly)
www.bristol.ac.uk/jobs/find/de...
Assistant Professor (L/SL) in AI, including computer vision [DL 5 Mar] @BristolUni - awarded AI University of the Year in 2024.
DM to myself or @_SethBullock_ for inquiries (don't us send CV pls!, apply directly)
www.bristol.ac.uk/jobs/find/de...
a cartoon pikachu says join us in white letters
ALT: a cartoon pikachu says join us in white letters
media.tenor.com
January 28, 2025 at 6:53 PM
Apply!
Assistant Professor (L/SL) in AI, including computer vision [DL 5 Mar] @BristolUni - awarded AI University of the Year in 2024.
DM to myself or @_SethBullock_ for inquiries (don't us send CV pls!, apply directly)
www.bristol.ac.uk/jobs/find/de...
Assistant Professor (L/SL) in AI, including computer vision [DL 5 Mar] @BristolUni - awarded AI University of the Year in 2024.
DM to myself or @_SethBullock_ for inquiries (don't us send CV pls!, apply directly)
www.bristol.ac.uk/jobs/find/de...
Reposted by Max Ilse
🎓 💫 We are opening post-doc positions at the intersection of AI, data science, and medicine:
• Large Language Models for French medical texts
• Evaluating digital medical devices: statistics and causal inference
• Large Language Models for French medical texts
• Evaluating digital medical devices: statistics and causal inference
January 29, 2025 at 8:19 AM
🎓 💫 We are opening post-doc positions at the intersection of AI, data science, and medicine:
• Large Language Models for French medical texts
• Evaluating digital medical devices: statistics and causal inference
• Large Language Models for French medical texts
• Evaluating digital medical devices: statistics and causal inference
Reposted by Max Ilse
If you name your AI benchmark "Humanity's Last Exam" and get Kevin Roose to gush about it, you work in advertising, not in computer science.
www.nytimes.com/2025/01/23/t...
www.nytimes.com/2025/01/23/t...
When A.I. Passes This Test, Look Out
The creators of a new test called “Humanity’s Last Exam” argue we may soon lose the ability to create tests hard enough for A.I. models.
www.nytimes.com
January 26, 2025 at 12:09 AM
If you name your AI benchmark "Humanity's Last Exam" and get Kevin Roose to gush about it, you work in advertising, not in computer science.
www.nytimes.com/2025/01/23/t...
www.nytimes.com/2025/01/23/t...
Reposted by Max Ilse
Happy to announce a collaboration with the Mayo Clinic to advance our research in radiology report generation!
newsnetwork.mayoclinic.org/discussion/m...
Tagging some of the core team: @valesalvatelli.bsky.social @fepegar.com @maxilse.bsky.social @sambondtaylor.bsky.social @anton-sc.bsky.social
newsnetwork.mayoclinic.org/discussion/m...
Tagging some of the core team: @valesalvatelli.bsky.social @fepegar.com @maxilse.bsky.social @sambondtaylor.bsky.social @anton-sc.bsky.social
newsnetwork.mayoclinic.org
January 15, 2025 at 10:56 AM
Happy to announce a collaboration with the Mayo Clinic to advance our research in radiology report generation!
newsnetwork.mayoclinic.org/discussion/m...
Tagging some of the core team: @valesalvatelli.bsky.social @fepegar.com @maxilse.bsky.social @sambondtaylor.bsky.social @anton-sc.bsky.social
newsnetwork.mayoclinic.org/discussion/m...
Tagging some of the core team: @valesalvatelli.bsky.social @fepegar.com @maxilse.bsky.social @sambondtaylor.bsky.social @anton-sc.bsky.social
Reposted by Max Ilse
The video of our talk "From Augustus to #Trump – Why #Disinformation Remains a Problem and What We Can Do About It Anyway" at #38c3, Europe's largest Hacker conference, was published, including the German original and English and Spanish translations:
media.ccc.de/v/38c3-von-a...
media.ccc.de/v/38c3-von-a...
On 2024-12-29 12:00, Josephine B. Schmitt and I will give a talk at Europe's largest Hacker conference #38C3:
“From Augustus to Trump – Why Disinformation Remains a Problem and What We Can Do About It Anyway”.
There will be live translation to English.
fahrplan.events.ccc.de/congress/202...
“From Augustus to Trump – Why Disinformation Remains a Problem and What We Can Do About It Anyway”.
There will be live translation to English.
fahrplan.events.ccc.de/congress/202...
January 10, 2025 at 8:49 AM
The video of our talk "From Augustus to #Trump – Why #Disinformation Remains a Problem and What We Can Do About It Anyway" at #38c3, Europe's largest Hacker conference, was published, including the German original and English and Spanish translations:
media.ccc.de/v/38c3-von-a...
media.ccc.de/v/38c3-von-a...
Reposted by Max Ilse
Internship in our group at Mila in reinforcement learning + graphs for reducing energy use in buildings.
More info and submit an application by Jan 13 here:
forms.gle/TCChXnvSAHqz...
Questions? Email donna.vakalis@mila.quebec with [intern!] in the subject line.
More info and submit an application by Jan 13 here:
forms.gle/TCChXnvSAHqz...
Questions? Email donna.vakalis@mila.quebec with [intern!] in the subject line.
January 10, 2025 at 1:14 AM
Internship in our group at Mila in reinforcement learning + graphs for reducing energy use in buildings.
More info and submit an application by Jan 13 here:
forms.gle/TCChXnvSAHqz...
Questions? Email donna.vakalis@mila.quebec with [intern!] in the subject line.
More info and submit an application by Jan 13 here:
forms.gle/TCChXnvSAHqz...
Questions? Email donna.vakalis@mila.quebec with [intern!] in the subject line.
Reposted by Max Ilse
Dear computer vision researchers, students & practitioners🔇🔇🔇
Remi Denton & I have written what I consider to be a comprehensive paper on the harms of computer vision systems reported to date & how people have proposed addressing them, from different angles.
PDF: cdn.sanity.io/files/wc2kmx...
Remi Denton & I have written what I consider to be a comprehensive paper on the harms of computer vision systems reported to date & how people have proposed addressing them, from different angles.
PDF: cdn.sanity.io/files/wc2kmx...
December 16, 2024 at 4:52 PM
Dear computer vision researchers, students & practitioners🔇🔇🔇
Remi Denton & I have written what I consider to be a comprehensive paper on the harms of computer vision systems reported to date & how people have proposed addressing them, from different angles.
PDF: cdn.sanity.io/files/wc2kmx...
Remi Denton & I have written what I consider to be a comprehensive paper on the harms of computer vision systems reported to date & how people have proposed addressing them, from different angles.
PDF: cdn.sanity.io/files/wc2kmx...
Reposted by Max Ilse
Help us commemorate victims, preserve memory & educate the world. Amplify our voice.
Your interaction here is more than just a click. It is an act of remembrance against forgetting. Like, share, or quote our posts.
Let people know that @auschwitzmemorial.bsky.social is present here.
Your interaction here is more than just a click. It is an act of remembrance against forgetting. Like, share, or quote our posts.
Let people know that @auschwitzmemorial.bsky.social is present here.
November 29, 2024 at 4:41 PM
Help us commemorate victims, preserve memory & educate the world. Amplify our voice.
Your interaction here is more than just a click. It is an act of remembrance against forgetting. Like, share, or quote our posts.
Let people know that @auschwitzmemorial.bsky.social is present here.
Your interaction here is more than just a click. It is an act of remembrance against forgetting. Like, share, or quote our posts.
Let people know that @auschwitzmemorial.bsky.social is present here.
Reposted by Max Ilse
New timeline, same problems, same solution
December 13, 2024 at 2:22 PM
New timeline, same problems, same solution
Kudos to @blackhc.bsky.social for calling out this neurips oral: openreview.net/forum?id=0NM... for not giving the rho-loss paper (arxiv.org/abs/2206.07137) the recognition it deserves!
December 12, 2024 at 10:14 AM
Kudos to @blackhc.bsky.social for calling out this neurips oral: openreview.net/forum?id=0NM... for not giving the rho-loss paper (arxiv.org/abs/2206.07137) the recognition it deserves!
Reposted by Max Ilse
In all of the reposts I see of articles criticising generative AI, I still don't see enough mention of work like Dr. Birhane's, which shows the biases against disadvantaged groups in the training datasets.
This is very good.
This is very good.
oh, this is out and they make me sound like an angel but a really good read
“We are not evaluating systems for some hypothetical, potential risks in the future,” Birhane says
www.fastcompany.com/91238006/how...
“We are not evaluating systems for some hypothetical, potential risks in the future,” Birhane says
www.fastcompany.com/91238006/how...
How Abeba Birhane is cleaning up AI’s dirty data
“We are not evaluating systems for some hypothetical, potential risks in the future. These audits are uncovering actual real issues, real problems, whether it's racism, sexism, or encoding of stereoty...
www.fastcompany.com
December 10, 2024 at 7:20 PM
In all of the reposts I see of articles criticising generative AI, I still don't see enough mention of work like Dr. Birhane's, which shows the biases against disadvantaged groups in the training datasets.
This is very good.
This is very good.
Reposted by Max Ilse
The next generation of probabilistic machine learning for weather called GenCast is published in @natureportfolio.bsky.social today 🥳. Amazing to see the collective progress in ML for weather as a field over the last 5 years. 🏖️ www.nature.com/articles/s41...
Probabilistic weather forecasting with machine learning - Nature
GenCast, a probabilistic weather model using artificial intelligence for weather forecasting, has greater skill and speed than the top operational medium-range weather forecast in the world and provid...
www.nature.com
December 4, 2024 at 7:30 PM
The next generation of probabilistic machine learning for weather called GenCast is published in @natureportfolio.bsky.social today 🥳. Amazing to see the collective progress in ML for weather as a field over the last 5 years. 🏖️ www.nature.com/articles/s41...
Reposted by Max Ilse
One of the reasons the university sector has come so spectacularly off the rails is the fact it's so unfriendly to family life, people with caring responsibilities and parents. The attitude is often: 'Not working 24/7? You're not fully committed!'
www.science.org/content/arti...
www.science.org/content/arti...
November 29, 2024 at 9:10 AM
One of the reasons the university sector has come so spectacularly off the rails is the fact it's so unfriendly to family life, people with caring responsibilities and parents. The attitude is often: 'Not working 24/7? You're not fully committed!'
www.science.org/content/arti...
www.science.org/content/arti...
Reposted by Max Ilse
Medically adapted foundation models (think Med-*) turn out to be more hot air than hot stuff. Correcting for fatal flaws in evaluation, the current crop are no better on balance than generic foundation models, even on the very tasks for which benefits are claimed.
arxiv.org/abs/2411.04118
arxiv.org/abs/2411.04118
Medical Adaptation of Large Language and Vision-Language Models: Are We Making Progress?
Several recent works seek to develop foundation models specifically for medical applications, adapting general-purpose large language models (LLMs) and vision-language models (VLMs) via continued pret...
arxiv.org
November 26, 2024 at 6:12 PM
Medically adapted foundation models (think Med-*) turn out to be more hot air than hot stuff. Correcting for fatal flaws in evaluation, the current crop are no better on balance than generic foundation models, even on the very tasks for which benefits are claimed.
arxiv.org/abs/2411.04118
arxiv.org/abs/2411.04118
Reposted by Max Ilse
What a surprise (not!). Yet again ... poor evaluations of specialized medical LLMs result in overhyped claims relative to the base LLMs. #bioMLeval
Medically adapted foundation models (think Med-*) turn out to be more hot air than hot stuff. Correcting for fatal flaws in evaluation, the current crop are no better on balance than generic foundation models, even on the very tasks for which benefits are claimed.
arxiv.org/abs/2411.04118
arxiv.org/abs/2411.04118
Medical Adaptation of Large Language and Vision-Language Models: Are We Making Progress?
Several recent works seek to develop foundation models specifically for medical applications, adapting general-purpose large language models (LLMs) and vision-language models (VLMs) via continued pret...
arxiv.org
November 27, 2024 at 2:16 AM
What a surprise (not!). Yet again ... poor evaluations of specialized medical LLMs result in overhyped claims relative to the base LLMs. #bioMLeval
Reposted by Max Ilse
Even as an interpretable ML researcher, I wasn't sure what to make of Mechanistic Interpretability, which seemed to come out of nowhere not too long ago.
But then I found the paper "Mechanistic?" by
@nsaphra.bsky.social and @sarah-nlp.bsky.social, which clarified things.
But then I found the paper "Mechanistic?" by
@nsaphra.bsky.social and @sarah-nlp.bsky.social, which clarified things.
November 20, 2024 at 8:00 AM
Even as an interpretable ML researcher, I wasn't sure what to make of Mechanistic Interpretability, which seemed to come out of nowhere not too long ago.
But then I found the paper "Mechanistic?" by
@nsaphra.bsky.social and @sarah-nlp.bsky.social, which clarified things.
But then I found the paper "Mechanistic?" by
@nsaphra.bsky.social and @sarah-nlp.bsky.social, which clarified things.
Reposted by Max Ilse
I can't find African Scientists and researchers easily on here. As the second largest continent with 1.3 billion inhabitants our perspectives matter. I'm creating a starterpack to find folks, follow them and amplify them. If you are an African Scientist let me know below so I can add you. Thanks.
November 20, 2024 at 11:08 AM
I can't find African Scientists and researchers easily on here. As the second largest continent with 1.3 billion inhabitants our perspectives matter. I'm creating a starterpack to find folks, follow them and amplify them. If you are an African Scientist let me know below so I can add you. Thanks.
Following the example of @sedielem.bsky.social I will repost some content:
Our latest work is using image editing via diffusion to stress-test biomedical vision models.
paper: arxiv.org/abs/2312.12865
code and model weights: huggingface.co/microsoft/ra...
blog: www.microsoft.com/en-us/resear...
Our latest work is using image editing via diffusion to stress-test biomedical vision models.
paper: arxiv.org/abs/2312.12865
code and model weights: huggingface.co/microsoft/ra...
blog: www.microsoft.com/en-us/resear...
RadEdit: stress-testing biomedical vision models via diffusion image editing
Biomedical imaging datasets are often small and biased, meaning that real-world performance of predictive models can be substantially lower than expected from internal testing. This work proposes usin...
arxiv.org
November 20, 2024 at 11:30 AM
Following the example of @sedielem.bsky.social I will repost some content:
Our latest work is using image editing via diffusion to stress-test biomedical vision models.
paper: arxiv.org/abs/2312.12865
code and model weights: huggingface.co/microsoft/ra...
blog: www.microsoft.com/en-us/resear...
Our latest work is using image editing via diffusion to stress-test biomedical vision models.
paper: arxiv.org/abs/2312.12865
code and model weights: huggingface.co/microsoft/ra...
blog: www.microsoft.com/en-us/resear...
Reposted by Max Ilse
We recently published CausalPlayground, a python library for causal data generation that can easily integrate with gymnasium. I hope that this can be useful for many.
Check it out on GitHub github.com/sa-and/Causa...
Check it out on GitHub github.com/sa-and/Causa...
GitHub - sa-and/CausalPlayground: The CausalPlayground library serves as a tool for causality research, focusing on the interactive exploration of structural causal models (SCMs). It provides extensiv...
The CausalPlayground library serves as a tool for causality research, focusing on the interactive exploration of structural causal models (SCMs). It provides extensive functionality for creating, m...
github.com
November 19, 2024 at 3:04 PM
We recently published CausalPlayground, a python library for causal data generation that can easily integrate with gymnasium. I hope that this can be useful for many.
Check it out on GitHub github.com/sa-and/Causa...
Check it out on GitHub github.com/sa-and/Causa...