I propose a novel framework that integrates stochastic differential equations (SDEs) with deep generative models to improve uncertainty quantification in machine learning applications involving structured and temporal data. This approach, termed Stochastic Latent Differential Inference (SLDI), embeds an Itô SDE in the latent space of a variational autoencoder, allowing for flexible, continuous-time modeling of uncertainty while preserving a principled mathematical foundation. The drift and diffusion terms of the SDE are parameterized by neural networks, enabling data-driven inference and generalizing classical time series models to handle irregular sampling and complex dynamic structure. A central theoretical contribution is the co-parameterization of the adjoint state with a dedicated neural network, forming a coupled forward-backward system that captures not only latent evolution but also gradient dynamics. I introduce a pathwise-regularized adjoint loss and analyze variance-reduced gradient flows through the lens of stochastic calculus, offering new tools for improving training stability in deep latent SDEs. My paper unifies and extends variational inference, continuous-time generative modeling, and control-theoretic optimization, providing a rigorous foundation for future developments in stochastic probabilistic machine learning.

Building efficient and effective generative models for neural network weights has been a research focus of significant interest that faces challenges posed by the high-dimensional weight spaces of modern neural networks and their symmetries. Several prior generative models are limited to generating partial neural network weights, particularly for larger models, such as ResNet and ViT. Those that do generate complete weights struggle with generation speed or require finetuning of the generated models. In this work, we present DeepWeightFlow, a Flow Matching model that operates directly in weight space to generate diverse and high-accuracy neural network weights for a variety of architectures, neural network sizes, and data modalities. The neural networks generated by DeepWeightFlow do not require fine-tuning to perform well and can scale to large networks. We apply Git Re-Basin and TransFusion for neural network canonicalization in the context of generative weight models to account for the impact of neural network permutation symmetries and to improve generation efficiency for larger model sizes. The generated networks excel at transfer learning, and ensembles of hundreds of neural networks can be generated in minutes, far exceeding the efficiency of diffusion-based methods. DeepWeightFlow models pave the way for more efficient and scalable generation of diverse sets of neural networks.

This paper establishes convergence rates for learning elliptic pseudo-differential operators, a fundamental operator class in partial differential equations and mathematical physics. In a wavelet-Galerkin framework, we formulate learning over this class as a structured infinite-dimensional regression problem with multiscale sparsity. Building on this structure, we propose a sparse, data- and computation-efficient estimator, which leverages a novel matrix compression scheme tailored to the learning task and a nested-support strategy to balance approximation and estimation errors. In addition to obtaining convergence rates for the estimator, we show that the learned operator induces an efficient and stable Galerkin solver whose numerical error matches its statistical accuracy. Our results therefore contribute to bringing together operator learning, data-driven solvers, and wavelet methods in scientific computing.

Feature attribution is the dominant paradigm for explaining deep neural networks. However, most existing methods only loosely reflect the model's prediction-making process, thereby merely white-painting the black box. We argue that explanatory alignment is a key aspect of trustworthiness in prediction tasks: explanations must be directly linked to predictions, rather than serving as post-hoc rationalizations. We present model readability as a design principle enabling alignment, and PiNets as a modeling framework to pursue it in a deep learning context. PiNets are pseudo-linear networks that produce instance-wise linear predictions in an arbitrary feature space, making them linearly readable. We illustrate their use on image classification and segmentation tasks, demonstrating how PiNets produce explanations that are faithful across multiple criteria in addition to alignment.

The pivot doesn't look out of place at CES, where Chinese electronics companies are increasingly applying their manufacturing prowess to new industries. If you've never been to Shenzhen, China's electronics capital, the annual CES trade show in Las Vegas is the next best thing. I'm reporting this week from the sprawling event, surrounded by fancy, strange, and often unnecessary gadgets, and despite my sore legs, I've barely scratched the surface. There are at least 900 Chinese tech companies attending CES this year, almost a quarter of the total exhibitors, according to an analysis of the conference's exhibitor directory. I even saw two Chinese humanoid robots at different booths dancing to the same viral Chinese rap song five minutes apart.

Why Are Grok and X Still Available in App Stores? Elon Musk's chatbot has been used to generate thousands of sexualized images of adults and apparent minors. Apple and Google have removed other "nudify" apps--but continue to host X and Grok. Elon Musk's AI chatbot Grok is being used to flood X with thousands of sexualized images of adults and apparent minors wearing minimal clothing. Some of this content appears to not only violate X's own policies, which prohibit sharing illegal content such as child sexual abuse material (CSAM), but may also violate the guidelines of Apple's App Store and the Google Play store.

Will Your Favorite Apps Be Along for the Ride? Tech companies are calling AI the next platform. But some developers are reluctant to let AI agents stand between them and their users. Silicon Valley giants like Amazon, Meta, and OpenAI are racing to develop "operating systems" for AI-powered devices--and 2026 is likely the year these efforts will start to take off. The devices are largely built around a future where AI agents can take actions on a user's behalf, without requiring them to visit an app or website.

It's the first week of a new year and there's no time for the tech world to slowly ease back into things following the holidays. That's because CES 2026 is in full swing, with all manner of companies descending on Las Vegas to reveal their latest innovations and what they're planning to bring your way in the near future. Many of the Engadget crew are on the ground to check out as much of the new tech as possible. Samsung held its First Look presentation, which focuses on home products, while LG has shown off a wide array of TVs and Lego unveiled its new Smart Brick technology. We've heard from the major chipmakers, gone hands-on with Samsung's trifold phone, checked out some funky laptops and seen some cute robots. There's some hot gaming gear at the show too, not to mention some weird tech . You don't necessarily have to wait to get your hands on all of these gadgets either. Some are available to buy right now. You can catch up on all of the big CES 2026 announcements (and some of the more offbeat gizmos we've seen) right here. We'll be keeping this story updated throughout the week. We also have CES live updates, with all the latest news from the event. Lego introduced the Smart Brick at CES 2026. In its first CES appearance, Lego announced the Smart Brick, a standard-sized brick with a 4.1mm ASIC chip inside that's designed to respond in different ways depending on what set you're building and how you're building it. Using what Lego calls the "Play Engine" and integrated copper coils, each brick can sense things like motion, orientation and magnetic fields, plus its own distance, direction and orientation in relation to other Smart Bricks. Each brick also has a teeny tiny speaker built in that will play audio "tied to live play actions" rather than only pre-recorded clips. Accompanying Smart Bricks are Smart Tags and Smart Minifigures, which have their own capabilities -- one of which is letting Smart Bricks know what context they are being used in. All of these pieces tie together via a local wireless layer dubbed BrickNet that, in part, lets Smart Bricks know where they are placed in relation to other smart components.

The Download: mimicking pregnancy's first moments in a lab, and AI parameters explained Plus: Google and Character.AI have settled a lawsuit linking their AI to the death of a teenager At first glance, it looks like the start of a human pregnancy: A ball-shaped embryo presses into the lining of the uterus then grips tight, burrowing in as the first tendrils of a future placenta appear. This is implantation--the moment that pregnancy officially begins. Only none of it is happening inside a body. These images were captured in a Beijing laboratory, inside a microfluidic chip, as scientists watched the scene unfold. In three recent papers published by Cell Press, scientists report what they call the most accurate efforts yet to mimic the first moments of pregnancy in the lab. They've taken human embryos from IVF centers and let these merge with "organoids" made of endometrial cells, which form the lining of the uterus.

Google Is Adding an'AI Inbox' to Gmail That Summarizes Emails New Gmail features, powered by the Gemini model, are part of Google's continued push for users to incorporate AI into their daily life and conversations. Google is putting even more generative AI tools into Gmail as part of its goal to further personalize user inboxes and streamline searches. On Thursday, the company announced a new "AI Inbox" tab, currently in a beta testing phase, that reads every message in a user's Gmail and suggests a list of to-dos and key topics, based on what it summarizes . In Google's example of what this AI Inbox could look like in Gmail, the new tab takes context from a user's messages and suggests they reschedule their dentist appointment, reply to a request from their child's sports coach, and pay an upcoming fee before the deadline. Also under the AI Inbox tab is a list of important topics worth browsing, nestled beneath the action items at the top.