Trump's 50 percent Canada tariffs: What to know US President Donald Trump's administration has said it will expand a voluntary pledge seeking to shield consumers from the energy costs of the rapid expansion of data centres, mostly used by artificial intelligence companies. The White House announced on Thursday that it would add state governors and electricity companies to the agreement, first announced with tech and AI firms in March. But the US administration stopped short of any enforceable protections. The pledge is a "public commitment that hyperscalers, AI companies, and the utilities and data-center developers behind them will build, bring, or buy every kilowatt their facilities need -- and cover every dollar of the infrastructure that delivers it". It says consumers would not foot the bill for AI's energy needs.

Last month, more than a hundred Stanford students left their own graduation to protest Google's military contracts and deals with ICE. Two organizers, Amanda Campos and Eva Jones, tell us why. Are the kids all right? The answer to that question probably depends on your criteria. At the very least, I can tell you that some of them are most definitely very mad. The anger percolating among America's youth was on display this graduation season, as college students across the country booed and jeered several high-profile commencement speakers who dared discuss--let alone extol the benefits of-- artificial intelligence . While AI use among college kids is widespread, so too is the fear and mistrust about what the technology might mean for the future. A recent Gallup poll found that Gen Z sentiment toward AI has significantly declined, with only 22 percent of respondents expressing excitement about the tech--and anxiety around AI affecting 42 percent of them. Among the commencement speech chaos, though, one school stood out to me: Stanford University. The California-based institution, long a key hub for Silicon Valley recruiting and a pipeline into elite tech gigs, this year saw more than a hundred students walk out of Google CEO Sundar Pichai's remarks to graduates. Ironically, Pichai didn't explicitly mention AI in his address at all.

The Japanese parliament has approved a bill to relax imperial succession rules, amid concerns over the dwindling size of the imperial family. The bill, passed by the upper house on Friday, allows the imperial family to adopt distant male relatives over the age of 15 and lets women keep their royal status after marrying outside the family. But it does not change the law barring women from ascending the throne despite wide public support for a female emperor, meaning Princess Aiko, the only child of the current emperor, is still not eligible to succeed the throne. The bill cleared the lower house last week, and will move through the final legal procedures before the changes take effect. Japan has the world's oldest continuous hereditary monarchy, with a lineage that's believed to span more than 2,600 years.

More people are using generative artificial intelligence to make travel plans for their summer vacation and letting their children use the technology when doing their homework during summer holidays. Six out of 10 people who responded to a survey on this year's summer holidays said they are using generative artificial intelligence to make travel plans. The survey, conducted by Meiji Yasuda Life Insurance on 1,120 people in their 20s to 50s in June, showed that 61.2% of those planning to travel in Japan or abroad refer to generative AI to make travel itineraries, as well as obtain information on local food and transportation. "The main tool people use for planning trips and doing research when they get there is shifting from travel guidebooks to generative AI," the firm said. Asked how they plan to spend their summer holidays, 58.4% said they are going out, down by 6.3 percentage points from last year. The rate of those traveling in Japan was 57.6%, up by 1 percentage point, while the ratio of those traveling overseas halved from 13.5% last year to 6.4%.

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?

AI-assisted interviews promise to reduce respondent burden in surveys by allowing respondents to describe experiences naturally while an AI system noisily maps those accounts into structured survey variables. That mapping is a measurement process that is fallible, versioned, adaptive, and potentially behaves differently across subgroups. This paper proposes Adaptive Matrix Validation (AMV), a design in which each respondent completes an AI-assisted interview, which is then mapped into tabular data by the AI. Respondents are also asked a small, randomized set of structured questions, which are used for statistical adjustment. The estimator first calibrates the mapped values using validation answers from other respondents, then corrects the remaining error with the validation answers observed for the target respondent. The paper develops estimators for item means, subgroup estimates, and regression coefficients when outcomes, predictors, or both are mapped from interviews. It also gives planning formulas the number of validation questions required and the sample size. A design-calibration simulation, an American Time Use Survey emulation, and a CHAMPS verbal-autopsy narrative study show when sparse validation can improve precision and when it cannot

Emotion recognition from physiological signals has substantial potential for applications in mental health and emotion-aware systems.

While reinforcement learning (RL) over chains of thought has significantly advanced language models in tasks such as mathematics and coding, visual reasoning introduces added complexity by requiring models to direct visual attention, interpret perceptual inputs, and ground abstract reasoning in spatial evidence. We introduce ViGoRL (Visually Grounded Reinforcement Learning), a vision-language model trained with RL to explicitly anchor each reasoning step to specific visual coordinates. Inspired by human visual decision-making, ViGoRL learns to produce spatially grounded reasoning traces, guiding visual attention to task-relevant regions at each step. When fine-grained exploration is required, our novel multi-turn RL framework enables the model to dynamically zoom into predicted coordinates as reasoning unfolds. Across a diverse set of visual reasoning benchmarks--including SAT-2 and BLINK for spatial reasoning, V bench for visual search, and ScreenSpot and VisualWebArena for web-based grounding--ViGoRL consistently outperforms both supervised fine-tuning and conventional RL baselines that lack explicit grounding mechanisms. Incorporating multi-turn RL with zoomed-in visual feedback significantly improves ViGoRL's performance on localizing small GUI elements and visual search, achieving 86.4% on V Bench. Additionally, we find that grounding amplifies other visual behaviors such as region exploration, grounded subgoal setting, and visual verification. Finally, human evaluations show that the model's visual references are not only spatially accurate but also helpful for understanding model reasoning steps. Our results show that visually grounded RL is a strong paradigm for imbuing models with general-purpose visual reasoning.

E.1 Limitations and Future Work - **1 -- Definitely AI-Generated**: Clear and frequent artifacts (e.g., blurry faces or objects, unnatural movements, inconsistent lighting), distorted shapes, 5) Exclude actions or descriptions (e.g., 'adjusting', 'imitating').

Prompt Template: Generating QAPairs for Camera Motion (CM) Task You are a professional question designer focusing on temporal dynamics in videos, including camera movements, motions, activities, and interactions, rather than static content. You will receive detailed annotations about the temporal details of the entire video, with duration markers in parentheses after "camera_motion" and "motion_list". Based on these annotations, design 3 multiple-choice questions around the "Camera Motion" theme to test models' fine-grained video motion understanding, particularly: Understanding camera movement direction and focus changes in the video. Additionally, follow these question design guidelines: 1. If a video's "camera_motion" has only one element, such as "camera_motion": "static", or "camera_motion": "camera shaking (0-22)", skip this video and don't generate any content.