AI Language Models: Inevitable Vulnerabilities and New SafeguardsCDS and Meta AI researchers have shown that language model vulnerabilities are inevitable but have developed a new method to make them…1d ago1d ago
At the Frontier of Research Independence: The CDS Faculty Fellow PositionHear from two current Fellows on their thoughts on CDS’s Faculty Fellow position, which is accepting application till November 25.6d ago6d ago
Robustness at a Cost: New Research Reveals Hidden Challenges in AI SecurityCDS research reveals that achieving AI robustness may demand significantly more data due to implicit biases.Oct 30Oct 30
New Framework Improves Multi-Modal AI Performance Across Diverse TasksCDS researchers propose a framework that improves multi-modal AI performance across diverse tasks, including healthcare diagnostics.Oct 25Oct 25
Improving AI’s Ability to Detect New Particles: A Multi-Background Approach for Physics DiscoveriesCDS researchers introduce a novel AI method for detecting rare particles at the Large Hadron Collider.Oct 231Oct 231
From Academia to Industry: How a 2018 Paper Foreshadowed OpenAI’s Latest InnovationHow a 2018 paper by CDS researchers helped shape OpenAI’s latest innovation, the o1 model.Oct 16Oct 16
Extending Diffusion Models to Nonlinear Processes: A Leap Forward for Science and AICDS’ Rajesh Ranganath and Courant PhD students extend diffusion models to nonlinear processes, enabling new AI applications.Oct 11Oct 11
AI Accelerates Universe and Planet Formation Simulations, Bringing Origins into FocusCDS’ Shirley Ho and team’s AI models transform simulations of the universe and planetary evolution from months to minutes.Oct 9Oct 9
Meet the Faculty: Emily BlackEmily Black joins CDS, bringing expertise in responsible AI and a mission to develop fairer algorithmic systems.Oct 4Oct 4
Human Intelligence Still Outshines AI on Abstract Reasoning TasksHumans outperform AI on visual reasoning tasks, says a new study by NYU researchers.Oct 3Oct 3