Global Covariance Pooling (GCP) has garnered increasing attention in visual recognition tasks, where second-order statistics frequently yield stronger representations than first-order approaches.

Plus: Anthropic has released a safe version of Mythos. The "steroid olympics" were a circus--and a window into our culture A couple of weeks ago, at a $50 million arena built in a casino parking lot in Las Vegas, I witnessed a libertarian thought experiment come to life. The inaugural Enhanced Games were the first sporting competition where participants were encouraged to take performance-enhancing drugs. For supporters of the event, the Enhanced Games offered a glimpse of a future in which medical advances push the human race to new heights--and they never have to get old. As I watched the games unfold, two questions bounced around my head: were they right? And what does that mean for the rest of us?

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?

When you purchase through links in our articles, we may earn a small commission. Could private jets predict the end of the world? By monitoring more than 31,000 aircraft, this website looks for unusual private jet departures that might signal trouble ahead. The internet is full of curious projects, but Apocalypse Early Warning System might be one of the most curious. It takes a semi tongue-in-cheek approach to alerting us to the end of the world.

It's an interim measure while the European Commission investigates the ban. The European Union has ordered Meta to open WhatsApp to AI chatbots from rival companies again, for free, as it investigates the messaging app's owner over potential antitrust violations. Meta introduced a new policy in October 2025 that banned third-party AI chatbots from the WhatsApp for Business API, making Meta AI the only chatbot that can access the service. Before the ban, companies could send notifications through WhatsApp, such as order alerts, using other AI assistants. EU officials opened an antitrust investigation into the new policy in December and then warned the company earlier this year that it can take interim measures against it. In its announcement, the commission explained that Meta has held a dominant position in the European messaging app market since at least 2023.

Integrating large language models (LLMs) into recommender systems has created new opportunities for improving recommendation quality. However, a comprehensive benchmark is needed to thoroughly evaluate and compare the recommendation capabilities of LLMs with traditional recommender systems. In this paper, we introduce \recbench{}, which systematically investigates various item representation forms (including unique identifier, text, semantic embedding, and semantic identifier) and evaluates two primary recommendation tasks, i.e., click-through rate prediction (CTR) and sequential recommendation (SeqRec). Our extensive experiments cover up to 17 large models and are conducted across five diverse datasets from fashion, news, video, books, and music domains. Our findings indicate that LLM-based recommenders outperform conventional recommenders, achieving up to a 5% AUC improvement in CTR and up to a 170% NDCG@10 improvement in SeqRec. However, these substantial performance gains come at the expense of significantly reduced inference efficiency, rendering LLMs impractical as real-time recommenders. We have released our code and data to enable other researchers to reproduce and build upon our experimental results.

Recent years have witnessed substantial progress on monocular depth estimation, particularly as measured by the success of large models on standard benchmarks. However, performance on standard benchmarks does not offer a complete assessment, because most evaluate accuracy but not robustness. In this work, we introduce PDE (Procedural Depth Evaluation), a new benchmark which enables systematic evaluation of robustness to changes in 3D scene content. PDE uses procedural generation to create 3D scenes that test robustness to various controlled perturbations, including object, camera, material and lighting changes. Our analysis yields interesting findings on what perturbations are challenging for state-of-the-art depth models, which we hope will inform further research.

Reinforcement learning (RL) has achieved significant success across a wide range of domains, however, most existing methods are formulated in discrete time. In this work, we introduce a novel RL method for continuous-time control, where stochastic differential equations govern state-action dynamics. Departing from traditional value function-based approaches, our key contribution is the characterization of continuous-time Q-functions via a martingale condition and the linking of diffusion policy scores to the action gradient of a learned continuous Q-function by the dynamic programming principle.

Large reasoning models (LRMs) have demonstrated impressive capabilities in complex problem-solving, yet their internal reasoning mechanisms remain poorly understood. In this paper, we investigate the reasoning trajectories of LRMs from an information-theoretic perspective. By tracking how mutual information (MI) between intermediate representations and the correct answer evolves during LRM reasoning, we observe an interesting MI peaks phenomenon: the MI at specific generative steps exhibits a sudden and significant increase during LRM's reasoning process. We theoretically analyze such phenomenon and show that as MI increases, the probability of model's prediction error decreases. Furthermore, these MI peaks often correspond to tokens expressing reflection or transition, such as Hmm, Wait and Therefore, which we term as the thinking tokens. We then demonstrate that these thinking tokens are crucial for LRM's reasoning performance, while other tokens has minimal impacts. Building on these analyses, we propose two simple yet effective methods to improve LRM's reasoning performance, by delicately leveraging these thinking tokens. Overall, our work provides novel insights into the reasoning mechanisms of LRMs and offers practical ways to improve their reasoning capabilities.

Ukrainian forces have carried out a missile attack deep inside Russia, hitting a major military plant overnight, President Volodymyr Zelensky has said. He said FP-5 Flamingo cruise missiles struck the drone and missile plant in the city of Cheboksary, in the Chuvash Republic, more than 900km (560 miles) from the front line. Local officials said three people were injured in a missile attack on the city. Ukraine also said it had hit the Moscow-occupied port of Mariupol on the Sea of Azov, a Russian oil refinery in Samara and a shadow fleet oil tanker in the Black Sea. In recent months, Ukraine's military has intensified its drone strikes on key facilities across Russia.