WRLD

Aggregated public profile · Not an official account.

Marinka Zitnik

@marinkazitnik wrld.org
Slovenian computer scientist
Marinka Zitnik Oct 10
Slovenian computer scientist

The post announced new ToolUniverse features that it said could automatically set up GPU environments, allowing agents to run machine-learning models on complex biological datasets and use the results in discovery campaigns. It credited Gao Shanghua with the work, in collaboration with NVIDIA.

3037
Marinka Zitnik Sep 30
Slovenian computer scientist

The author praised new work by Google DeepMind on watermarking AI-generated proteins while preserving their function and design quality. They also said they had discussed watermarking and other ways to track generative AI outputs in biology with Science Magazine.

Marinka Zitnik Sep 1
Slovenian computer scientist

Closed-loop AI scientists AI can generate hypotheses much faster than we can experimentally verify them. @AdaFang_ In our latest piece on Closing the Loop in AI-Driven Biomedical Discovery we outline a path for closed-loop AI scientists, https://t.co/FBo1oniHWb AI scientists https://t.co/S2Ie5p1PYG #ClosedLoopAI #BiomedicalDiscovery #AIScience #Nospecificcountrycodesarementionedinthetweet.

118
Marinka Zitnik Jul 7
Slovenian computer scientist

@GaoShanghua @ayushnoori @CurtGinder @BenGlicksberg @RanBalicer @HarvardDBMI @KempnerInst @biohub @IcahnMountSinai @ClalitHealth @BrighamWomens 🟢 Benchmarking: Across 3,168 drug reasoning questions (DrugPC) and 456 patient specific treatment cases (TreatmentPC), ATHENA hit 94.7% and 82.9% accuracy, beating GPT-5 by 17.8 and 10.7 points and DeepSeek-R1 (671B) by even more. What stood out: giving GPT-5 optional tool https://t.co/xdmOHeiS8U #AI #DrugDevelopment #HealthcareInnovation #Nospecificcountriesarementionedinthetweet.

Marinka Zitnik Jul 7
Slovenian computer scientist

Introducing ATHENA: an AI agent for treatment reasoning across all FDA approved drugs since 1939, by stellar @GaoShanghua Paper: https://t.co/598JNOxWZ1 Code: https://t.co/aaGYtsVXKI Project: https://t.co/ntjZidLCo9 Every treatment decision means weighing disease context, https://t.co/ddiuVoiYFm #AI #Healthcare #DrugReform #Thetweetdoesnotmentionanyspecificcountries.Therefore #therearenocountrycodestoreturn.

Marinka Zitnik Jul 3
Slovenian computer scientist

New @NatureMedicine paper: A pan-cancer foundation model predicts immunotherapy response and highlights the biology behind it https://t.co/hfVQc24r6w @harvardmed news: https://t.co/1w6u8cEYBK Roughly a quarter of patients with advanced cancer respond durably to immune #CancerResearch #Immunotherapy #HealthTech #Nospecificcountrycodesfound.

239
Marinka Zitnik Jun 30
Slovenian computer scientist

Every AI action in medicine, logged. Check out MedLog in this new thread by @ayushnoori 👇 Share feedback, join our team and pilot MedLog at your institution, https://t.co/plEIFjCYww #MedTech #AIinHealthcare #MedLog #Thetweetdoesnotmentionanyspecificcountries.Hence #therearenocountrycodestoreturn.

358
Marinka Zitnik Jun 28
Slovenian computer scientist

How to log and audit models, agents, and AI systems in medicine? @ayushnoori MedLog is a syslog for medical AI. Medicine deploys AI on millions of patients with no systematic record of what these models do, when they fail, or who they benefit or harm most. No standard captures https://t.co/LBgWQgb6rP #MedicalAI #AIEthics #HealthTech #Therearenospecificcountriesmentionedinthetweet.