Depth Pro 2: Sharp and Temporally Consistent Metric Monocular Depth
We present Depth Pro 2, an approach to estimating dense metric depth for images and video at full resolution. By analyzing prior methods for monocular depth estimation, we find that they commonly trade off temporal consistency and boundary sharpness: multi-frame methods favor consistency, single-fra...
AI-Enabled Video System Monitors ICU Patients for Early Signs of Delirium
Computer vision experts have developed a 24/7 video monitoring system that tracks patient activity and helps doctors identify subtle signs of delirium.
Goemans's Unsplittable-Flow Conjecture: The Anatomy of AI-Assisted Counterexample Discovery
arXiv:2611.23244v1 Announce Type: new
Abstract: Goemans conjectured that any fractional single-source flow could be converted into an unsplittable flow without increasing total cost and while increasing any arc load by at most the largest demand. Recently, Dmitry Rybin announced a seven-vertex coun...
FACE: Flow-Guided Angular Channel Estimation Framework for mmWave MIMO Systems
arXiv:2611.21698v1 Announce Type: new
Abstract: Acquiring accurate channel state information (CSI) is essential for millimeter-wave (mmWave) multi-input multi-output (MIMO) systems. However, it remains challenging due to high-dimensional channels, limited pilot observations, and measurement noise. ...
At Stanford HAI’s fall conference, experts explored how decisions about governance, work, and education can turn AI’s promise into broader public benefit.
Scientists tracked kids for 8 years. What they found about screen time was surprising
A surprising Finnish study found that children who spent more time using screens tended to have better cognitive processing abilities as teenagers, challenging the assumption that screen time is always harmful. Researchers suggest that what children do on screens may matter more than how much time t...
D\'ecoupage: Directorial Audio-Visual Control for Long Multi-Shot Video Generation
arXiv:2611.18866v1 Announce Type: new
Abstract: In cinematic storytelling, sound and visual transitions do not always coincide in lockstep: dialogue frequently bridges across cuts to reveal reactions or introduces new scenes via J- and L-cuts, while deliberate silences pace the scene. Existing audi...
Multimodal Fusion: A Review of Breast Cancer Advanced and Traditional Modalities
arXiv:2611.18793v1 Announce Type: new
Abstract: Purpose: Multimodal medical imaging fusion (MMIF) is an increasingly important approach to breast cancer detection (BCD), as different imaging modalities rely on different physical principles and provide complementary information. No single modality c...
How to make a moral agent: sentimentalism and rationalism as lenses on AI
arXiv:2611.18714v1 Announce Type: new
Abstract: Can AI systems be moral agents? That is, can they undertake actions that can be correctly assessed as morally right or wrong? There is no settled normative criterion of moral agency that applies neatly outside paradigm human cases. Because of the diff...
Quantum computers are getting too powerful to check. Scientists have a solution
Scientists have developed a new way to determine whether quantum simulations can be trusted, solving a growing problem as quantum computers become too powerful for conventional machines to check. Demonstrated on a 51-qubit system, the technique measures uncertainty and provides verifiable error marg...
Classification with Large-Step Gradient Descent: Unifying Self-Stabilization and Convergence Rates
arXiv:2611.17459v1 Announce Type: new
Abstract: Neural networks are often trained with gradient descent (GD) using step sizes that exceed classical stability thresholds, a regime where standard convergence theory no longer applies. Despite exhibiting non-monotonic loss and oscillatory dynamics, thi...
Alignment with Reflect-Reasoning: Learning How To Follow Principles
arXiv:2611.17398v1 Announce Type: new
Abstract: The appropriate governing norms and principles for the alignment of Large Language Models (LLMs) often depend on cultural context and deployment application. However, most alignment techniques focus on training models to conform to human preferences o...
We’re shortening the disclosure window in our coordinated vulnerability disclosure (CVD) policy from 90 days to 30 days for high- and critical-severity issues, and to 45 days for all other issues.
Stanford HAI Backs Ambitious AI Projects with 2026 Hoffman-Yee Grants
Seven interdisciplinary research teams win a total of $3.5 million to pursue bold human-centered approaches to AI across medicine, science, law and education.
AI discovers a hidden heart signal that could predict sudden death
Scientists have uncovered a hidden warning sign in routine heart scans that could predict sudden cardiac death before it strikes. The AI-powered discovery outperformed conventional screening methods, potentially revealing thousands of people at risk of a fatal heart malfunction that doctors might ot...
New findings from the Stanford AI Index reveal a more uneven picture of public sentiment and labor market outcomes than previously thought, as AI investment surpasses $1 trillion.
Zugwise: Domain-Aware Game Reconstruction for Digitizing Handwritten Chess Scoresheets
arXiv:2611.14544v1 Announce Type: new
Abstract: Digitizing handwritten chess scoresheets has been studied as a handwriting-recognition problem: read the move in each box and check that it is legal. This framing misses the errors that matter most. A misread that yields an illegal move is loud. One t...
AI-powered bionic eye breakthrough could help the blind see again
Scientists have developed an AI-powered approach that makes experimental bionic vision technology more precise, helping predict the flashes and shapes of light perceived by a blind man with a brain implant. By learning how an individual's brain responds to electrical stimulation, the technology coul...