Soroosh Tayebi Arasteh
PulseAugur coverage of Soroosh Tayebi Arasteh — every cluster mentioning Soroosh Tayebi Arasteh across labs, papers, and developer communities, ranked by signal.
1 day(s) with sentiment data
-
New framework analyzes privacy impact on medical image AI
Researchers have introduced a new framework called Differential Privacy Representation Geometry for Medical Imaging (DP-RGMI) to better understand how differential privacy affects medical image analysis. This framework …
-
New FRAMEwork Assesses AI Fairness Bias in Medical Imaging
Researchers have developed a new framework called FRAME (Fair-model Reference And Mechanism Evaluation) to better assess fairness bias in medical imaging AI models. FRAME distinguishes between performance differences ca…
-
Pretraining domain crucial for private medical image AI, study finds
A new study published on arXiv investigates the impact of pretraining data on the privacy-utility trade-off in medical image analysis models. Researchers found that the domain of the pretraining data, specifically using…
-
Self-supervision, not clinical data, drives medical AI model convergence
A new study published on arXiv investigates the convergence of representations in medical foundation models. Researchers found that self-supervised learning objectives drive this convergence more significantly than clin…
-
New CANDOR measure reveals weaknesses in frozen AI encoders
Researchers have developed CANDOR, a new discordance measure designed to more accurately assess the capabilities of frozen foundation encoders. Unlike previous methods, CANDOR uses symmetric, equal-sized banks to fix it…
-
LLMs retain clinical evidence strength signals internally but fail to express them
A new study reveals that large language models (LLMs) can internally represent the strength of clinical evidence, even when they fail to express this confidence in their stated grades. Researchers found that a linear es…
-
New benchmarks and models advance vision-language capabilities in robotics and reasoning · 10 sources tracked
Recent research explores advancements in vision-language models (VLMs) across several domains. DeCAL introduces a new model for dexterous manipulation that integrates tactile sensing and visual-language understanding. R…