Why Iran won't give up Hormuz Could Israel sabotage US-Iran deal? Lawmakers in the United States are quietly advancing a proposal that could deepen military ties between the US and Israel in unprecedented ways, at a time when public support for Israel among Americans is increasingly fractured. Among the provisions included in the 2027 National Defence Authorisation Act (NDAA) released this week is Section 224, the "United States-Israel Defence Technology Cooperation Initiative". Some legislators have already signalled opposition, with Representative Thomas Massie saying he would seek to remove the provision if it reaches the House floor. The measure remains at an early stage, but analysts say if passed, it would limit political oversight over the defence relationship.

History-dependent sampling can reduce long-run Monte Carlo variance by discouraging redundant revisits, but existing schemes typically encode history through empirical measure on finite state spaces, which is infeasible in high-dimensional discrete configuration spaces or ill-posed in continuous domains. We propose Score-Repellent Monte Carlo (SRMC) framework that summarizes trajectory history by a running average of score evaluations in $\mathbb{R}^d$, where $d$ is the dimension of the score and state representation. This history is converted into a surrogate target through an exponential score tilt, indexed with $α$ that represents the strength of repellence in controlling the magnitude of the history-based repulsion. The surrogate family is normalization-free in the standard MCMC sense, yielding a generic wrapper: at each iteration, any base kernel targeting $π$ can instead be run on the current surrogate $π_{θ_n}$ while the history is updated online. We analyze the coupled evolution of the history recursion and Monte Carlo estimators using stochastic approximation with controlled Markovian noise, establishing almost sure convergence and a joint central limit theorem. We further identify regimes in which the asymptotic covariance decreases as $α$ increases, with scaling $O(1/α)$, extending the near-zero-variance effect of finite-state history-dependent samplers to general state spaces with constant memory. Experiments on continuous targets and discrete energy-based models demonstrate improved estimator variance and mode coverage, while retaining $O(d)$ memory usage and modest per-iteration overhead.

We study inference-time alignment for diffusion-based generative models, aiming to steer a base model toward high-reward outputs without updating its weights. Recent Sequential Monte Carlo (SMC)-based steering methods approximate reward-tilted target distributions in a principled way, but their proposals remain largely tied to the base sampler. Since reward information is mainly used after propagation through particle reweighting and resampling, these methods can require large particle budgets and suffer from weight degeneracy and high-variance estimates. One way to reduce variance and improve particle efficiency is to iteratively learn twisting functions that provide look-ahead guidance, as in twisted SMC. However, existing learnable twisting methods are developed mainly for classical sequential inference and can be unstable when applied to diffusion-based alignment with high-dimensional state spaces and terminal, noisy, or black-box rewards. We propose Trust-Region Iterative Twisted Sequential Monte Carlo (TRI-TSMC), a trust-region framework for learning twisting functions in SMC-based inference-time alignment. Each iteration computes an exact KL-constrained update in path space, which admits a closed-form solution by tempered importance reweighting, and projects this target back to the parameterized twisted family by weighted maximum likelihood. Theoretically, we formalize the value-function interpretation of the optimal twisting function and show that it yields a zero-variance sampler. We prove that the trust-region update follows an escort path toward the target distribution, that the weighted maximum-likelihood update is a forward-KL projection, and that the path reduces residual importance-weight variance. Empirically, TRI-TSMC improves primary alignment objectives on discrete diffusion text generation and text-to-image generation under matched inference-time budgets.

RFK Jr. and Mehmet Oz's comments about teen sperm count and "underbabied" Americans at a recent women's health event underscore the White House's pronatalist agenda. Robert F. Kennedy Jr., secretary of Health and Human Services, and President Donald Trump discuss workplace IVF benefits on October 16, 2025, in Washington, DC. The home page for Moms.gov, the Trump administration's recently launched website for "new and expecting mothers," is a trad wife's dream. Featuring soft pastel graphics and a photo of a young, white, blond woman in a field clutching her pregnant belly, the website offers resources for women of reproductive age such as anti-abortion "pregnancy centers," as well as a CDC website listing potential workplace hazards for expecting mothers without noting accompanying legal protections for pregnant women. If you were conspiratorially minded, you might conclude from the website alone that the Trump administration is champing at the bit for young (white and blond) women to have as many (white and blond) babies as possible.

GameStop has built up a stake of 5% in eBay and is offering to acquire the company at $125 a share. GameStop has built up a stake of 5% in eBay and is offering to acquire the company at $125 a share. GameStop's $55.5bn bid for eBay rejected as'neither credible nor attractive' Online marketplace takes into account uncertainty around US video games retailer's financing proposal The board of eBay has rejected the US video games retailer GameStop's surprise $55.5bn bid (£41bn) for the online marketplace, describing the proposal as "neither credible nor attractive". Earlier this month, GameStop made an unsolicited bid for eBay, publishing a letter on its website outlining a half-cash, half-stock proposal. This was despite the US games company - which became a global household name during the meme stock craze of 2021 - being worth far less than its takeover target.

California gubernatorial candidate Tom Steyer is proposing a new jobs guarantee for workers displaced by artificial intelligence. Billionaire California gubernatorial candidate Tom Steyer is rolling out a new proposal that would guarantee jobs with benefits for workers displaced by artificial intelligence . He's the first state-wide candidate to make such a pledge. The plan, which builds on a broader AI policy framework Steyer released in March, promises to make California "the first major economy in the world" to ensure "good-paying" jobs to workers impacted by AI. To do so, Steyer tells WIRED he plans to build off a previous proposal to introduce a "token tax" which would tax big tech companies "a fraction of a cent for every unit of data processed" for AI.

GameStop's CEO said he could turn eBay into something worth hundreds of billions of dollars. GameStop's CEO said he could turn eBay into something worth hundreds of billions of dollars. GameStop makes $55.5bn takeover offer for eBay Video game retailer's CEO warns that unsolicited bid could turn hostile if it is rebuffed by resale site's board US video games retailer GameStop has offered to buy eBay for $55.5bn (£41bn) in an unsolicited bid that its boss warned could turn hostile if the proposal is rebuffed by eBay's board. GameStop, which has quietly accumulated a 5% stake in eBay, said it was willing to pay $125 a share, split 50-50 between cash and stock. It is an ambitious move by the games company, which catapulted to fame during the meme-stock craze of 2021 but is worth far less than its takeover target.

Sequential Monte Carlo (SMC) is an inference algorithm for state space models that approximates the posterior by sampling from a sequence of target distributions. The target distributions are often chosen to be the filtering distributions, but these ignore information from future observations, leading to practical and theoretical limitations in inference and model learning. We introduce SIXO, a method that instead learns target distributions that approximate the smoothing distributions, incorporating information from all observations. The key idea is to use density ratio estimation to fit functions that warp the filtering distributions into the smoothing distributions. We then use SMC with these learned targets to define a variational objective for model and proposal learning. SIXO yields provably tighter log marginal lower bounds and offers more accurate posterior inferences and parameter estimates in a variety of domains.