Variational mean field approximations tend to struggle with contemporary overparametrized deep neural networks. Where a Bayesian treatment is usually associated with high-quality predictions and uncertainties, the practical reality has been the opposite, with unstable training, poor predictive power, and subpar calibration. Building upon recent work on reparametrizations of neural networks, we propose a simple variational family that considers two independent linear subspaces of the parameter space. These represent functional changes inside and outside the support of training data. This allows us to build a fully-correlated approximate posterior reflecting the overparametrization that tunes easy-to-interpret hyperparameters. We develop scalable numerical routines that maximize the associated evidence lower bound (ELBO) and sample from the approximate posterior. Empirically, we observe state-of-the-art performance across tasks, models, and datasets compared to a wide array of baseline methods. Our results show that approximate Bayesian inference applied to deep neural networks is far from a lost cause when constructing inference mechanisms that reflect the geometry of reparametrizations.

Outside of the Utah hospital where Charlie Kirk was taken Wednesday, a crowd of families, college students, and Proud Boys gathered to mourn. Mourners outside Timpanogos Regional Hospital in Orem, Utah, where Charlie Kirk was taken after being shot at Utah Valley University, September 10, 2025. At around 8 pm on Wednesday, the motorcade with Charlie Kirk's body left the Timpanogos Regional Hospital in Orem, Utah. Along the road, the 100 or so people who showed up for an impromptu memorial for Kirk stopped what they were doing, lined the sidewalk, and stared as it sped away from the mountains and into the dark. Kirk had been at nearby Utah Valley University for the kickoff stop of his planned "American Comeback Tour" when a sniper allegedly stood on the roof of a building 200 or so yards away, and took aim, killing him with one shot .

A vishing attack is a form of social engineering where attackers use phone calls to deceive individuals into disclosing sensitive information, such as personal data, financial information, or security credentials. Attackers exploit the perceived urgency and authenticity of voice communication to manipulate victims, often posing as legitimate entities like banks or tech support. Vishing is a particularly serious threat as it bypasses security controls designed to protect information. In this work, we study the potential for vishing attacks to escalate with the advent of AI. In theory, AI-powered software bots may have the ability to automate these attacks by initiating conversations with potential victims via phone calls and deceiving them into disclosing sensitive information. To validate this thesis, we introduce ViKing, an AI-powered vishing system developed using publicly available AI technology. It relies on a Large Language Model (LLM) as its core cognitive processor to steer conversations with victims, complemented by a pipeline of speech-to-text and text-to-speech modules that facilitate audio-text conversion in phone calls. Through a controlled social experiment involving 240 participants, we discovered that ViKing has successfully persuaded many participants to reveal sensitive information, even those who had been explicitly warned about the risk of vishing campaigns. Interactions with ViKing's bots were generally considered realistic. From these findings, we conclude that tools like ViKing may already be accessible to potential malicious actors, while also serving as an invaluable resource for cyber awareness programs.

'Gutfeld!' panelists react to Google pausing its image generation feature of its artificial intelligence (AI) tool, Gemini, after AI refuses to show images of White people. Save the energy for after the show. Can goo goo goo goo, can Google be trusted when their credibility is busted? Google's apologizing after their new AI Gemini chat bot created historically inaccurate pictures and refusing to show White people. For those unfamiliar with the software, you describe what you want to see and AI generates the images.

But Google's Gemini has been accused of being racist towards white people. The tool uses artificial intelligence to create images from prompts within seconds. But users claim the AI bot refuses to create images of Caucasian people, after testing it with requests for Popes, Vikings, and country music fans. 'New game: Try to get Google Gemini to make an image of a Caucasian male. I have not been successful so far,' one user wrote on X (formerly Twitter).

Robotic navigation has been approached as a problem of 3D reconstruction and planning, as well as an end-to-end learning problem. However, long-range navigation requires both planning and reasoning about local traversability, as well as being able to utilize general knowledge about global geography, in the form of a roadmap, GPS, or other side information providing important cues. In this work, we propose an approach that integrates learning and planning, and can utilize side information such as schematic roadmaps, satellite maps and GPS coordinates as a planning heuristic, without relying on them being accurate. Our method, ViKiNG, incorporates a local traversability model, which looks at the robot's current camera observation and a potential subgoal to infer how easily that subgoal can be reached, as well as a heuristic model, which looks at overhead maps for hints and attempts to evaluate the appropriateness of these subgoals in order to reach the goal. These models are used by a heuristic planner to identify the best waypoint in order to reach the final destination. Our method performs no explicit geometric reconstruction, utilizing only a topological representation of the environment. Despite having never seen trajectories longer than 80 meters in its training dataset, ViKiNG can leverage its image-based learned controller and goal-directed heuristic to navigate to goals up to 3 kilometers away in previously unseen environments, and exhibit complex behaviors such as probing potential paths and backtracking when they are found to be non-viable. ViKiNG is also robust to unreliable maps and GPS, since the low-level controller ultimately makes decisions based on egocentric image observations, using maps only as planning heuristics. For videos of our experiments, please check out our project page https://sites.google.com/view/viking-release.

MailOnline has tried out'AI Time Machine', an online tool that can transform you into a Viking, a Greek warrior, an Egyptian pharaoh or even a 1960s hippy. The new feature from MyHeritage reimagines any adult as if they were from another historical era, simply using a small sample of uploaded photos. The science team fed photos of James Corden, Piers Morgan and Drew Barrymore into the tool – and got some rather hilarious results. Corden looks well-groomed and ready for battle in his Viking clobber, while Morgan makes a real mean-looking bandit from the Wild West. Piers Morgan appears here as a Roman empire legionary.

We consider the problem of time series forecasting in an adaptive setting. We focus on the inference of state-space models under unknown and potentially time-varying noise variances. We introduce an augmented model in which the variances are represented as auxiliary gaussian latent variables in a tracking mode. As variances are nonnegative, a transformation is chosen and applied to these latent variables. The inference relies on the online variational Bayesian methodology, which consists in minimizing a Kullback-Leibler divergence at each time step. We observe that the minimum of the Kullback-Leibler divergence is an extension of the Kalman filter taking into account the variance uncertainty. We design a novel algorithm, named Viking, using these optimal recursive updates. For auxiliary latent variables, we use second-order bounds whose optimum admit closed-form solutions. Experiments on synthetic data show that Viking behaves well and is robust to misspecification.

Your partner asks you why the little, evil dudes in a certain game called Overlord speak as if they were stolen from a Monty Python sketch. Your terse response--being an evil Overlord while commanding a horde of unruly minions is hard goddamn work, after all--is that someone was paid a good amount of money to make them sound that way. But the question sticks in your mind as the in-game banter continues to amuse, so much so that you find yourself laughing out loud. As the credits roll, you make sure to note the person responsible for the quips and barbs: Rhianna Pratchett. After a quick Google search, you find that she's the daughter of the famous Discworld author Terry Pratchett, and that she began as a gaming journalist before crossing over to write for games rather than about them.

"Valhalla" recognizes the beauty and unique perspectives we each bring to the stories we've known and loved throughout history. By refusing to show us a "true" depiction of the Isu civilization, and viewing it only through the lens of cultures we know, we begin to have a deeper connection to the past, even as the truth remains forever obscured from us. Stories convey history, culture and most importantly, the values we inherit and pass on in traditions and teachings. There are stories being created every time we forge alliances, as well as during times of rebellion, disagreement and discord. Stories strengthen the bonds of community, or can widen the fissures between them.