I'm almost embarrassed to admit that the first time I saw one of Meta's new holographic avatars today, during a WhatsApp call with a Meta rep I had just met a few minutes before, I didn't immediately clock that I was speaking to an avatar. That's how far the company's avatar technology has come in the last couple years. Meta announced at Connect that it's bringing its holographic avatars to its new VR glasses and to the Meta Ray-Ban Display frames. I got the chance to experience both ahead of the keynote and was surprised at how far the tech has come since the first time I experienced it on Meta's Orion prototype. Setting up an avatar is a bit different depending on which device you're using.

More information Adding us as a Preferred Source in Google by using this link indicates that you would like to see more of our content in Google News results. We do look different in photos versus a mirror. If you're anything like me, you've had this experience: You see a photo of yourself--maybe on a day you actually felt relatively put-together before leaving the house--and you suddenly feel betrayed by your own face. For me, it's usually some combination of the following: one eye squinting against the sun, a smile that somehow always turns into a tortured rictus, my chin multiplying despite my best "strategic selfie " angle. So, if the relatively inoffensive face I saw in the mirror is me, and the deeply unfortunate JPEG is also me, which do I believe?

Follow this section to personalize your feed and get instant alerts. Follow Go to your personalized feed WHY FOLLOW? Smart Alerts: Get notified about major news as it happens. Follow this tag to personalize your feed and get instant alerts. Follow Go to your personalized feed WHY FOLLOW?

Dogs can tell if you're angry, scared, or sad Our canine companions may really understand what we're going through, new study suggests. More information Adding us as a Preferred Source in Google by using this link indicates that you would like to see more of our content in Google News results. Four canine study participants--Morante, Kun-kun, Odin, and Molly--are helping scientists understand more about how dogs recognize human emotions. Breakthroughs, discoveries, and DIY tips sent six days a week. By signing up, you confirm you are 16+, will receive newsletters and promotional content and agree to our Terms of Use and acknowledge the data practices in our Privacy Policy .

Multi-task regression aims at jointly solving multiple regression problems, called tasks. Compared to solving each task separately, better performances can be achieved as long as the tasks are sufficiently related. Full-conformal prediction is a framework that formulates a data-dependent prediction-region containing the unknown output-vector at any prescribed confidence level. However, explicit computation of this prediction-region is intractable in general since it requires training infinitely many predictors. The present work focuses on multi-task regression in a Reproducing Kernel Hilbert Space (RKHS) of vector-valued functions. This computational issue is addressed by designing an approximating predictionregion containing the full-conformal one. This construction is carried out in two scenarios: piq when the inter-task covariance-matrix is known, and piiq when this matrix is estimated. In terms of volume, the tightness of this approximation is assessed theoretically by means of an upper-bound in the first scenario. It is also empirically proved to improve upon the split-conformal prediction on synthetic data in both scenarios.

What Happened to Your Face? How the human countenance became something to study, edit, optimize, and scan. The physiognomists promised that your character could be read from your features. Certain forms of facial-recognition technology have revived that old fantasy in digital form. Several months ago, my partner and I bought an apartment in South London. Our previous home was a rental in which, for reasons best known to the landlord, there were mirrors everywhere. The bathroom had two; there was one outside on the terrace; in the bedroom, mirrored panels stretched across a twenty-foot-long wall. On moving day, we realized that we had a problem: the new apartment was mirror-free, and because we'd been so spoiled we weren't bringing one of our own. We spent a few days filling our drafty rooms, decanting books, building furniture, and dressing every morning without seeing ourselves in profile. It was a couple of weeks before we bought a simple mirror, wooden and round, to hang above the bathroom sink. By then, I joked, we didn't recognize ourselves.

Multi-agent systems (MAS) are central to applications such as swarm robotics and traffic routing, where agents must coordinate in a decentralized manner to achieve a common objective. Stochastic Shortest Path (SSP) problems provide a natural framework for modeling decentralized control in such settings. While the problem of learning in SSP has been extensively studied in single-agent settings, the decentralized multi-agent variant remains largely unexplored. In this work, we take a step towards addressing that gap. We study decentralized multi-agent SSPs (Dec-MASSPs) under linear function approximation, where the transition dynamics and costs are represented using linear models. Applying novel symmetry-based arguments, we identify the structure of optimal policies. Our main contribution is the first regret lower bound for this setting based on the construction of hard-tolearn instances for any number of agents, n. Our regret lower bound of Ω( K), over K episodes, highlights the inherent learning difficulty in Dec-MASSPs. These insights clarify the learning complexity of decentralized control and can further guide the design of efficient learning algorithms in multi-agent systems.

We introduce LoRAShop, the first framework for multi-concept image editing with LoRA models. LoRAShop builds on a key observation about the feature interaction patterns inside Flux-style diffusion transformers: concept-specific transformer features activate spatially coherent regions early in the denoising process. We harness this observation to derive a disentangled latent mask for each concept in a prior forward pass and blend the corresponding LoRA weights only within regions bounding the concepts to be personalized.

Diffusion LLMs have emerged as a promising alternative to conventional autoregressive LLMs, offering substantial potential for improving runtime efficiency. However, existing diffusion models fail to provably enforce user-specified formal constraints, such as regular expressions, which makes them unreliable for tasks that require structured outputs, such as fixed-schema JSON generation. Unlike autoregressive models, which generate tokens sequentially, diffusion LLMs predict a block of tokens in parallel. This parallelism makes traditional constrained decoding algorithms, designed to enforce constraints with sequential token prediction, ineffective at preserving the true output distribution. To address this limitation, we propose DINGO, a dynamic programming-based constrained decoding strategy that is both efficient and provably distribution-preserving. DINGO enables sampling of output strings with the highest probability under the model's predicted distribution while strictly adhering to any user-specified regular expression. On standard symbolic math and JSON generation benchmarks, DINGO achieves up to a 68%points of improvement over unconstrained inference. The code is available at DINGO.

What property of the data distribution determines the excess risk of principal component analysis? In this paper, we provide a precise answer to this question. We establish a central limit theorem for the error of the principal subspace estimated by PCA, and derive the asymptotic distribution of its excess risk under the reconstruction loss. We obtain a non-asymptotic upper bound on the excess risk of PCA that recovers, in the large sample limit, our asymptotic characterization. Underlying our contributions is the following result: we prove that the negative block Rayleigh quotient, defined on the Grassmannian, is generalized self-concordant along geodesics emanating from its minimizer of maximum rotation less than π/4.