Irregular multivariate time series impose a trade-off for long-horizon forecasting: discrete methods can distort temporal structure via re-gridding, while continuous-time models often require sequential solvers prone to drift. To bridge this gap, we present Latent Laplace Diffusion (LLapDiff), a generative framework that models the target as a low-dimensional latent trajectory, enabling horizon-wide generation without step-by-step integration over physical time. We guide the reverse process utilizing a stable modal parameterization motivated by stochastic port-Hamiltonian dynamics, and parameterize its mean evolution in the Laplace domain via learnable complex-conjugate poles, enabling direct evaluation over irregular timestamps. We also link continuous dynamics to irregular observations through renewal-averaging analysis, which maps sampling gaps to effective event-domain poles and motivates a gap-aware history summarizer. Extensive experiments show that LLapDiff improves over baselines in long-horizon forecasting, and its continuous-time generative nature supports missing-value imputation by querying the same model at historical timestamps. Code is available at https://github.com/pixelhero98/LLapDiffusion.

We introduce FLUXtrapolation, a benchmark for extrapolating ecosystem fluxes under progressively harder distribution shifts. Ecosystem fluxes are central to understanding the carbon, water, and energy cycles, yet they can only be measured directly at sparsely located measurement towers. Producing global flux estimates therefore requires training models on observed sites using globally available covariates and predicting in unobserved regions, that is, upscaling. Flux upscaling is a challenging domain generalization problem that is affected by a shift in covariate distribution across climates, ecosystem types, and environmental conditions, as well as by conditional shift: important drivers remain unobserved at global scale. We provide a quantitative analysis of both these shifts in $P_X$ and $P_{Y\mid X}$. FLUXtrapolation is designed based on domain expertise on flux upscaling: it defines temporal, spatial, and temperature-based extrapolation scenarios and evaluates performance across held-out domains, temporal aggregations, and tail errors. In a pilot study, we find that baselines perform similarly under median hourly RMSE, but separate under the proposed tail-focused and multi-scale evaluation. FLUXtrapolation therefore poses a realistic and thus relevant challenge for machine learning methods under distribution shift; at the same time, progress on this benchmark would directly support the scientific goal of improving flux upscaling.

Standard generative models struggle with heavy-tailed data: Lipschitz architectures cannot produce power-law tails from Gaussian noise, and interpolating between heavy-tailed data and Gaussians is ill-posed. We propose a simple fix: apply the soft-log transform $ϕ(x) = \mathrm{sign}(x) \cdot \log(1 + |x|)$ coordinate-wise to data before training, then exponentiate samples after generation. A Hill diagnostic decides per-coordinate whether to transform, leaving light-tailed margins untouched at no added complexity. This compresses heavy tails into a range where standard flow matching succeeds, without heavy-tailed base distributions or architectural modifications. We provide theoretical intuition for why this works: the log-transform maps Pareto tails to exponentials, and the induced dynamics implement a form of tail annealing via power transformations. On a 144-configuration multivariate benchmark (3 copulas, $d$ up to 100, 4 tail indices), Log-FM dominates specialized baselines on $W_1$, CVaR$_{99}$, and extreme-quantile metrics, and is the only method with zero severe divergences across 2{,}880 runs.

Bayesian optimization (BO) selects evaluation points for expensive black-box objectives using Gaussian process (GP) predictive distributions. Kernel choice and hyperparameter selection can lead to miscalibrated predictive distributions and an inappropriate exploration-exploitation trade-off. For minimization, sampling criteria such as expected improvement (EI) depend on the predictive distribution below the current best value, so lower-tail miscalibration directly affects the sampling decision. This article studies goal-oriented calibration of GP predictive distributions below a low threshold $t$ in the noiseless setting, for standard GP models with hyperparameters selected by maximum likelihood. A framework for predictive reliability below $t$ is introduced, based on two notions of spatial calibration: occurrence calibration over the design space and thresholded $μ$-calibration on sublevel sets of the form $\{x\in\mathbb{X}, f(x)\le t\}$. Building on this framework, we propose tcGP, a post-hoc method that calibrates GP predictive distributions below~$t$, and we show that the resulting EI-based global optimization algorithm remains dense in the design space. Experiments on standard benchmarks show improved lower-tail calibration and BO performance relative to standard GP models and globally calibrated GP models.

Israeli air strikes have killed at least 19 people in southern Lebanon, the country's health ministry has said. Ten of them, including three children and three women, were killed in a single attack that hit a house in the town of Deir Qanoun, the ministry said. Lebanon was drawn into the war on 2 March, when the Iran-backed armed Shia Islamist group Hezbollah fired rockets at Israel in retaliation for US-Israeli strikes that killed Iran's supreme leader. The latest deaths less than a week after the US said that Lebanon and Israel had agreed to extend a ceasefire by 45 days, with the two sides set to resume talks at the beginning of June. Despite the extension, both Israel and Hezbollah have continued to exchange fire, especially in southern Lebanon.

Gear Wearables Google's Android XR smart glasses hope to succeed where AI-first wearables have failed The audio-only frames pair with Android and iOS so a Gemini agent can run errands on your phone while you stay heads-up. 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 may earn revenue from the products available on this page and participate in affiliate programs. Google put AI on people's faces more than a decade ago with its Google Glass wearable. It was designed to put a computer directly on your face, but the world (and to some extent, the hardware) wasn't quite ready for that yet.

What are Russia's gains from the Iran war? 'We are not losers; we are winners' More than 15,800 people killed in Russia's all-out war on Ukraine: UN The United Nations has said 15,850 people, including 791 children, have been killed in Ukraine since Russia's full-scale invasion of the neighbouring country in February 2022. The "actual figures are likely significantly higher", Kayoko Gotoh, Europe and Central Asia director of the UN's Department of Political and Peacebuilding Affairs (DPPA), told the UN Security Council on Tuesday. US President Donald Trump has attempted to mediate and announced the most recent three-day ceasefire earlier this month, but fighting has resumed. Tuesday's Russian attacks on Ukraine killed at least six people. A 15-year-old boy was among three people killed in a Russian ballistic missile attack on the city of Pryluky in north-central Ukraine's Chernihiv region on Tuesday morning, according to the State Emergency Service of Ukraine.

Google is sprucing up its Gemini models, revamping search, and enabling AI agents in everything. There are also some spiffy new smart glasses coming this fall. Google just wrapped its keynote address at its annual I/O developer event . The company showed off a swath of new agentic AI features and some demos of its upcoming Android-powered smart glasses. As it has in the past few years, the spectacle largely revolved around Google's perpetual stream of AI efforts.

The Commonwealth Foundation said all entrants to the prize had avowed that their submissions were their own work. The Commonwealth Foundation said all entrants to the prize had avowed that their submissions were their own work. 'Obvious markers of AI': doubts raised over winner of short story prize Granta publisher says'perhaps we never will know' true authorship of work that won Commonwealth prize A few syntactical tics - and the verdict of an AI detection platform - have sparked a furore over the possibility that a short story given a prestigious literary award was written by AI. The foundation that awarded the prize and Granta, the magazine that published the winning story, said they had considered the allegations but had not reached a conclusion as to whether they were true. "It may be that the judges have now awarded a prize to an instance of AI plagiarism - we don't yet know, and perhaps we never will know," the publisher of Granta, Sigrid Rausing, said.

Estonia has said a Nato fighter jet shot down a drone, which it suspects was a Ukrainian projectile knocked off course by Russian electronic jamming, over its territory. Defence Minister Hanno Pevkur said a Romanian F-16 fired a missile and drone debris fell in a marshy area in central Estonia on Tuesday. Ukraine reacted by accusing Russia of deliberately redirecting Ukrainian drones launched at legitimate military targets in Russia, apologising to Estonia and all of our Baltic friends for such unintended incidents. Russia has not commented on the latest in a series of recent drone incursions over Nato members Estonia, Latvia and Lithuania. Last week, Latvian Prime Minister Evika Silina resigned following a political crisis over Russia-bound Ukrainian drones straying into Latvian territory.