Lightnews — Scholar-powered news

Christian Bluethgen

@cxbln.bsky.social

470 followers 280 following 49 posts

Radiologist @ USZ Zurich | Med AI research @ Stanford AIMI | #RadSky #ChestRad 🫁🫀 🩻

Posts Media Videos Starter Packs

Pinned

Christian Bluethgen @cxbln.bsky.social · Nov 23

A small starter pack for accounts in the radiology/AI space (or prolific in either of those fields) - happy to add more! go.bsky.app/9Drtasz #RadSky #Radiology #AI

11 11 27

Reposted by Christian Bluethgen

IAMJB @iamjbd.bsky.social · Jun 9

💥 We unveil our paper accepted at the #ACL2025 Main Conference:
Automated Structured Report Generation

Let's revisit automated radiology report generation for CXR.
Free-form reports make it hard for AI systems to learn accurate generation, and even harder to evaluate. 🧵👇
@StanfordAIMI @hopprai

1 3 7

Reposted by Christian Bluethgen

Woojin Kim, MD @woojinkim.com · May 5

👍 If you're interested in #LLMs in #radiology, this is a recommended read!

💯 While the article focuses primarily on LLMs, as the authors recommended, "Keep an eye on [large multimodal models]".

👉 pubs.rsna.org/doi/10.1148/...
#RadiologyAI #AIStrategy #LMMs

1 4

Reposted by Christian Bluethgen

Ethan Mollick @emollick.bsky.social · May 4

When reading AI benchmarks, aside from the fact that many of the AIs are (accidentally or on purpose) trained on the test set, many tests are just bad. MMLU likely maxes out at 90% or so because so many of the questions in it are just wrong. It is also uncalibrated in difficulty across questions.

3 27

Christian Bluethgen @cxbln.bsky.social · Apr 3

LLMs formally pass the Turing test

arxiv.org/abs/2503.23674

Large Language Models Pass the Turing Test

We evaluated 4 systems (ELIZA, GPT-4o, LLaMa-3.1-405B, and GPT-4.5) in two randomised, controlled, and pre-registered Turing tests on independent populations. Participants had 5 minute conversations s...

arxiv.org

Christian Bluethgen @cxbln.bsky.social · Apr 1

lot of folks talking about MCP and developers rushing to support it - but it does seem a bit like hype-y.

interesting perspective: hardcoresoftware.learningbyshipping.com/p/230-mcp-it...

230. MCP - It's Hot, But Will It Win?

There's a long history of "middleware" in our industry. Everyone wants it. There's always a hot one, but it rarely makes to the finish line and often disappoints.

hardcoresoftware.learningbyshipping.com

Christian Bluethgen @cxbln.bsky.social · Mar 12

that's a first :O

1 1

Christian Bluethgen @cxbln.bsky.social · Mar 10

arxiv.org/abs/2502.19634

MedVLM-R1: Incentivizing Medical Reasoning Capability of Vision-Language Models (VLMs) via Reinforcement Learning

Reasoning is a critical frontier for advancing medical image analysis, where transparency and trustworthiness play a central role in both clinician trust and regulatory approval. Although Medical Visu...

arxiv.org

1 2

Reposted by Christian Bluethgen

TechCrunch @techcrunch.com · Mar 6

A quarter of startups in YC’s current cohort have codebases that are almost entirely AI-generated

With the release of new AI models that are better at coding, developers are increasingly using AI to generate code. One of the newest examples is the current batch of Y Combinator, the storied Silicon Valley startup accelerator. A quarter of the W25…

tcrn.ch

3 6 16

Reposted by Christian Bluethgen

Tugba Akinci D'Antonoli @tugbaakinci.bsky.social · Mar 1

Chaired two insightful sessions at #ECR2025 today!

"Standardization and Reporting in AI Research" with experts Hans Reitsma Mike Klontzas Annika Reinke

"How to Use ChatGPT for Academic and Administrative Tasks" Andreas S. Brendlin @cxbln.bsky.social and Ghizlane Lembarki

@myesr.bsky.social

1 6

Christian Bluethgen @cxbln.bsky.social · Feb 23

still thinking about this interaction with Claude (Oct '24) from time to time

Christian Bluethgen @cxbln.bsky.social · Feb 5

r1ing away on an ordinary machine

.. straight out of Kahnemans lesser known book "Thinking .. mostly slow"

Christian Bluethgen @cxbln.bsky.social · Feb 5

For Germans unable to vote locally, it is easy to partake, here's how: www.bundesregierung.de/breg-de/schw...

Wahlwissen: Briefwahl | Bundesregierung

Wer am Wahltag zur Bundestagswahl 2025 verhindert ist, kann vorab per Briefwahl wählen. Alle wichtigen Informationen zur Briefwahl im Überblick.

www.bundesregierung.de

Christian Bluethgen @cxbln.bsky.social · Feb 2

Reposted by Christian Bluethgen

Jay Alammar @jayalammar.bsky.social · Jan 27

The Illustrated DeepSeek-R1

Spent the weekend reading the paper and sorting through the intuitions. Here's a visual guide and the main intuitions to understand the model and the process that created it.

newsletter.languagemodels.co/p/the-illust...

1 22 72

Christian Bluethgen @cxbln.bsky.social · Jan 22

it's a journey

Christian Bluethgen @cxbln.bsky.social · Jan 19

how’d you guess?

Christian Bluethgen @cxbln.bsky.social · Jan 7

another excellent blog post by @chiphuyen.bsky.social, this time on #agents

huyenchip.com/2025/01/07/a...

Agents

Intelligent agents are considered by many to be the ultimate goal of AI. The classic book by Stuart Russell and Peter Norvig, Artificial Intelligence: A Modern Approach (Prentice Hall, 1995), defines ...

huyenchip.com

Reposted by Christian Bluethgen

Gaël Varoquaux @gaelvaroquaux.bsky.social · Dec 19

People: please don't ML for the sake of ML.

I keep seeing manuscripts using fancy machine learning on brain-imaging data, where, in my opinion (having processed a lot of brain-imaging data), the method is way too complex for the richness of the data.

Fancier is not better per se

5 15 89

Christian Bluethgen @cxbln.bsky.social · Dec 26

my wl:
- fewer low-hanging proof-of-concept studies (yes, it works for your specialty/organ/classification, too)
- fewer head-to-head model comparisons (yes, llama 3.1 was better than llama 2 for your case, but peer review took 8 months, now there’s llama 4)

- instead: RCTs, relevant outcomes

1 2 8

Reposted by Christian Bluethgen

TechCrunch @techcrunch.com · Dec 24

The promise and perils of synthetic data

Is it possible for an AI to be trained just on data generated by another AI? It might sound like a harebrained idea. But it’s one that’s been around for quite some time — and as new, real data is increasingly hard to come by, it’s been gaining traction.…

tcrn.ch

1 12 28

Christian Bluethgen @cxbln.bsky.social · Dec 22

thanks for sharing

Christian Bluethgen @cxbln.bsky.social · Dec 22

"productization requires [standardization in development and deployment], which is antithetical to research. [...] PhDs are supposed to come up with innovative ideas, validate these ideas, report the findings to the community by writing papers and then move on."

[slightly edited to fit post limits]

kyunghyuncho.bsky.social @kyunghyuncho.bsky.social · Dec 21

feeling a but under the weather this week … thus an increased level of activity on social media and blog: kyunghyuncho.me/i-sensed-anx...

i sensed anxiety and frustration at NeurIPS’24 – Kyunghyun Cho

kyunghyuncho.me

Christian Bluethgen @cxbln.bsky.social · Dec 18

A market research team's dream: gathering real-world use cases and improvement opportunities directly from natural product usage

from: www.anthropic.com/research/clio

Reposted by Christian Bluethgen

Maarten van Smeden @maartenvsmeden.bsky.social · Dec 16

NEW PREPRINT

A detailed overview of 32 popular predictive performance metrics for prediction models

arxiv.org/abs/2412.10288

11 65 190

Christian Bluethgen @cxbln.bsky.social · Dec 13