On the 30th floor of the Chartered Square building in downtown Bangkok, the low-key office of Skyhub Technologies serves as a nexus for a burgeoning and contentious trade. The space, rented out by a serviced office provider, is visited only rarely by the company's sole director and occasionally by Chinese nationals, according to building staff who asked not to be identified speaking about clients. No contact number is listed on its online registration documents. No one was available during a visit in late January. Despite the appearance of inactivity, this is a busy conduit for advanced drones. Trade documents show that Skyhub Technologies is Thailand's second-biggest importer of unmanned aerial vehicles from China.

Learning to Defer (L2D) enables a classifier to abstain from predictions and defer to an expert, and has recently been extended to multi-expert settings. In this work, we show that multi-expert L2D is fundamentally more challenging than the single-expert case. With multiple experts, the classifier's underfitting becomes inherent, which seriously degrades prediction performance, whereas in the single-expert setting it arises only under specific conditions. We theoretically reveal that this stems from an intrinsic expert identifiability issue: learning which expert to trust from a diverse pool, a problem absent in the single-expert case and renders existing underfitting remedies failed. To tackle this issue, we propose PiCCE (Pick the Confident and Correct Expert), a surrogate-based method that adaptively identifies a reliable expert based on empirical evidence. PiCCE effectively reduces multi-expert L2D to a single-expert-like learning problem, thereby resolving multi expert underfitting. We further prove its statistical consistency and ability to recover class probabilities and expert accuracies. Extensive experiments across diverse settings, including real-world expert scenarios, validate our theoretical results and demonstrate improved performance.

The problem of domain generalization concerns learning predictive models that are robust to distribution shifts when deployed in new, previously unseen environments. Existing methods typically require labeled data from multiple training environments, limiting their applicability when labeled data are scarce. In this work, we study domain generalization in an anti-causal setting, where the outcome causes the observed covariates. Under this structure, environment perturbations that affect the covariates do not propagate to the outcome, which motivates regularizing the model's sensitivity to these perturbations. Crucially, estimating these perturbation directions does not require labels, enabling us to leverage unlabeled data from multiple environments. We propose two methods that penalize the model's sensitivity to variations in the mean and covariance of the covariates across environments, respectively, and prove that these methods have worst-case optimality guarantees under certain classes of environments. Finally, we demonstrate the empirical performance of our approach on a controlled physical system and a physiological signal dataset.

The proliferation of Large Language Models (LLMs) necessitates efficient mechanisms to distinguish machine-generated content from human text. While statistical watermarking has emerged as a promising solution, existing methods suffer from two critical limitations: the lack of a principled approach for selecting sampling distributions and the reliance on fixed-horizon hypothesis testing, which precludes valid early stopping. In this paper, we bridge this gap by developing the first e-value-based watermarking framework, Anchored E-Watermarking, that unifies optimal sampling with anytime-valid inference. Unlike traditional approaches where optional stopping invalidates Type-I error guarantees, our framework enables valid, anytime-inference by constructing a test supermartingale for the detection process. By leveraging an anchor distribution to approximate the target model, we characterize the optimal e-value with respect to the worst-case log-growth rate and derive the optimal expected stopping time. Our theoretical claims are substantiated by simulations and evaluations on established benchmarks, showing that our framework can significantly enhance sample efficiency, reducing the average token budget required for detection by 13-15% relative to state-of-the-art baselines.

A new artificial intelligence (AI) model developed by the Chinese company behind TikTok rocked Hollywood this week - not just because of what it can do, but what it could mean for creative industries. Created by tech giant ByteDance, Seedance 2.0 can generate cinema-quality video, complete with sound effects and dialogue, from just a few written prompts. Many of the clips said to have been made using Seedance, and featuring popular characters like Spider-Man and Deadpool, went viral. What is Seedance - and why the stir? Seedance was launched to little fanfare in June 2025 but it is the second version that came eight months later that has caused a major stir.

The United States trade deficit has widened sharply in December amid a surge in imports, and the goods shortfall in 2025 was the highest on record despite US President Donald Trump's tariffs on foreign-manufactured merchandise. The second straight monthly deterioration in the trade deficit reported by the US Commerce Department on Thursday suggested that trade made little or no contribution to gross domestic product (GDP) in the fourth quarter. The US deficit in the trade of goods widened 2 percent to a record $1.24 trillion last year as American companies boosted imports of computer chips and other tech goods from Taiwan to support massive investments in artificial intelligence. Amid continuing tensions with Beijing, the deficit in the goods trade with China plunged nearly 32 percent to $202bn in 2025 on a sharp drop in both exports to and imports from the world's second-biggest economy. But trade was diverted away from China.

Donald Trump Jr.'s Private DC Club Has Mysterious Ties to an Ex-Cop With a Controversial Past The Executive Branch has a reported membership list that includes Trumpworld elites like David Sacks. A WIRED review of corporate filings reveals an under-the-radar player: a notorious former DC police officer. When the Executive Branch soft-launched in Washington, DC, last spring, the private club's initial buzz centered on its starry roster of backers and founding members. The president's eldest son, Donald Trump Jr., is one of the club's several co-owners, according to previous reporting. Founding members reportedly include Trump administration AI czar David Sacks and his podcast cohost Chamath Palihapitiya, as well as crypto bigwigs Tyler and Cameron Winklevoss.

Neural Information Processing Systems http://nips.cc/