Large language models (LLMs) are increasingly proposed for detecting and responding to violent content online, yet their ability to reason about morally ambiguous, real-world scenarios remains underexamined. We present the first study to evaluate LLMs using a validated social science instrument designed to measure human response to everyday conflict, namely the Violent Behavior Vignette Questionnaire (VBVQ). To assess potential bias, we introduce persona-based prompting that varies race, age, and geographic identity within the United States. Six LLMs developed across different geopolitical and organizational contexts are evaluated under a unified zero-shot setting. Our study reveals two key findings: (1) LLMs surface-level text generation often diverges from their internal preference for violent responses; (2) their violent tendencies vary across demographics, frequently contradicting established findings in criminology, social science, and psychology.

DaCapo is a specialized deep learning library tailored to expedite the training and application of existing machine learning approaches on large, near-isotropic image data. In this correspondence, we introduce DaCapo's unique features optimized for this specific domain, highlighting its modular structure, efficient experiment management tools, and scalable deployment capabilities. We discuss its potential to improve access to large-scale, isotropic image segmentation and invite the community to explore and contribute to this open-source initiative.

Fox News' Martha MacCallum has the latest on her new Fox Nation documentary on'The Story.' My father-in-law passed away last month, days away from his 99th birthday. He lived with us for 13 years. He was a great man, a World War II veteran who loved his wife and raised three children. As his vascular dementia worsened – unlike Alzheimer's, his long-term memory remained intact almost until the end – my wife would set him up with a familiar film. "The Godfather" played most frequently, followed by "Patton."

Large skew-t factor copula models are attractive for the modeling of financial data because they allow for asymmetric and extreme tail dependence. We show that the copula implicit in the skew-t distribution of Azzalini and Capitanio (2003) allows for a higher level of pairwise asymmetric dependence than two popular alternative skew-t copulas. Estimation of this copula in high dimensions is challenging, and we propose a fast and accurate Bayesian variational inference (VI) approach to do so. The method uses a conditionally Gaussian generative representation of the skew-t distribution to define an augmented posterior that can be approximated accurately. A fast stochastic gradient ascent algorithm is used to solve the variational optimization. The new methodology is used to estimate copula models for intraday returns from 2017 to 2021 on 93 U.S. equities. The copula captures substantial heterogeneity in asymmetric dependence over equity pairs, in addition to the variability in pairwise correlations. We show that intraday predictive densities from the skew-t copula are more accurate than from some other copula models, while portfolio selection strategies based on the estimated pairwise tail dependencies improve performance relative to the benchmark index.

Managing a complex illness often requires different treatment regimens spread over a long time. The complexity of these potentially life-threatening diagnoses can be daunting to patients while they are most vulnerable. We present a vision for artificial intelligence-enabled tools for assisting patients in managing the complex information given to them over the course of their treatments through the combination of existing and emerging techniques from natural language processing and knowledge representation. We provide examples from an actual breast cancer diagnosis and treatment plan and highlight the development of new combinations of techniques to build tools that can reason about data from a variety of sources and act as intelligence-augmenting agents. We conclude with a discussion of some additional challenges facing artificial intelligence practitioners as applications become patient-centric.

Hit play on the player above to hear the podcast and follow along with the transcript below. This transcript was automatically generated, and then edited for clarity in its current form. There may be some differences between the audio and the text. Welcome back to Talking Tech. Brett Molina is off today.

Under pressure from U.S. auto safety regulators, Tesla has agreed to stop allowing video games to be played on center touch screens while its vehicles are moving. The National Highway Traffic Safety Administration says the company will send out a software update over the Internet so the function called "Passenger Play" will be locked and won't work while vehicles are in motion. The move comes one day after the agency announced it would open a formal investigation into distracted driving concerns about Tesla's video games, some of which could be played while cars are being driven. An agency spokeswoman says in a statement Thursday that the change came after regulators discussed concerns about the system with Tesla. The first update went out Wednesday as part of Tesla's holiday software release, and the rest of the vehicles should get it today.

Tesla claims the Model S Plaid is the world's quickest car and that the $129,990 sedan can accelerate to 60 mph in less than two seconds. Under pressure from U.S. auto safety regulators, Tesla has agreed to stop allowing video games to be played on center touch screens while its vehicles are moving. "Passenger Play" allowed video games to be played on front screen with vehicle in motion. The National Highway Traffic Safety Administration says the company will send out a software update over the Internet so the function called "Passenger Play" will be locked and won't work while vehicles are in motion. The move comes one day after the agency announced it would open a formal investigation into distracted driving concerns about Tesla's video games, some of which could be played while cars are being driven.

The US has opened a formal investigation into a report that Tesla vehicles allow people to play video games on a center touch screen while they are driving. The investigation by the National Highway Traffic Safety Administration covers about 580,000 electric cars and SUVs from model years 2017 through 2022. The action follows a complaint to the agency that Teslas equipped with "gameplay functionality" allow games to be played by the driver while in motion. In a document posted online Wednesday, the agency says the feature, called "Passenger Play," may distract the driver and increase the risk of a crash. The agency's office of defects investigation said in the document that the game capability has been available since December 2020.

A new Tesla owner demonstrates earlier this month on a closed course in Portland, Ore., how he can play video games on the vehicle's console while driving. The U.S. has opened a formal investigation into a report that Tesla vehicles allow people to play video games on a center touch screen while they are driving. A new Tesla owner demonstrates earlier this month on a closed course in Portland, Ore., how he can play video games on the vehicle's console while driving. The U.S. has opened a formal investigation into a report that Tesla vehicles allow people to play video games on a center touch screen while they are driving. ATHENS, Ohio -- The U.S. has opened a formal investigation into the potential for Tesla drivers to play video games on a center touch screen while the vehicle is in motion.