In the month since the new season of South Park began airing, the infamous animated show has somehow become even more depraved, and I mean that as a compliment. In Wednesday night's episode, "Sickofancy," creators Matt Stone and Trey Parker continue their all-out satirical assault on the Trump administration, and once again the series takes no prisoners. At one point, South Park's version of President Donald Trump suggests inserting a (very real) trophy gifted to him by Apple CEO Tim Cook up the anus of Trump's boyfriend, Satan. By the episode's end, we even see one longtime character forced into a new role as the president's "cum rag." Hey, if I had to be subjected to this image, you do too.

Nate Soares doesn't set aside money for his 401(k). "I just don't expect the world to be around," he told me earlier this summer from his office at the Machine Intelligence Research Institute, where he is the president. A few weeks earlier, I'd heard a similar rationale from Dan Hendrycks, the director of the Center for AI Safety. By the time he could tap into any retirement funds, Hendrycks anticipates a world in which "everything is fully automated," he told me. That is, "if we're around."

Most existing illumination-editing approaches fail to simultaneously provide customized control of light effects and preserve content integrity. This makes them less effective for practical lighting stylization requirements, especially in the challenging task of transferring complex light effects from a reference image to a user-specified target image. To address this problem, we propose TransLight, a novel framework that enables high-fidelity and high-freedom transfer of light effects. Extracting the light effect from the reference image is the most critical and challenging step in our method. The difficulty lies in the complex geometric structure features embedded in light effects that are highly coupled with content in real-world scenarios. To achieve this, we first present Generative Decoupling, where two fine-tuned diffusion models are used to accurately separate image content and light effects, generating a newly curated, million-scale dataset of image-content-light triplets. Then, we employ IC-Light as the generative model and train our model with our triplets, injecting the reference lighting image as an additional conditioning signal.The resulting TransLight model enables customized and natural transfer of diverse light effects. Notably, by thoroughly disentangling light effects from reference images, our generative decoupling strategy endows TransLight with highly flexible illumination control.Experimental results establish TransLight as the first method to successfully transfer light effects across disparate images, delivering more customized illumination control than existing techniques and charting new directions for research in illumination harmonization and editing.

Based on comparable English-Chinese news corpora annotated by Large Language Model (LLM), this paper attempts to explore the differences in constituent order of English-Chinese news from the perspective of functional chunks with adverbial roles, and analyze their typical positional preferences and distribution patterns. It is found that: (1) English news prefers linear narrative of core information first, and functional chunks are mostly post-positioned, while Chinese news prefers overall presentation mode of background first, and functional chunks are often pre-positioned; (2) In SVO structure, both English and Chinese news show differences in the distribution of functional chunks, but the tendency of Chinese pre-positioning is more significant, while that of English post-positioning is relatively mild; (3) When function blocks are co-occurring, both English and Chinese news show high flexibility, and the order adjustment is driven by information and pragmatic purposes. The study reveals that word order has both systematic preference and dynamic adaptability, providing new empirical support for contrastive study of English-Chinese information structure.

Current AI systems minimize risk by enforcing ideological neutrality, yet this may introduce automation bias by suppressing cognitive engagement in human decision-making. We conducted randomized trials with 2,500 participants to test whether culturally biased AI enhances human decision-making. Participants interacted with politically diverse GPT-4o variants on information evaluation tasks. Partisan AI assistants enhanced human performance, increased engagement, and reduced evaluative bias compared to non-biased counterparts, with amplified benefits when participants encountered opposing views. These gains carried a trust penalty: participants underappreciated biased AI and overcredited neutral systems. Exposing participants to two AIs whose biases flanked human perspectives closed the perception-performance gap. These findings complicate conventional wisdom about AI neutrality, suggesting that strategic integration of diverse cultural biases may foster improved and resilient human decision-making.

Current OCR systems are based on deep learning models trained on large amounts of data. Although they have shown some ability to generalize to unseen data, especially in detection tasks, they can struggle with recognizing low-quality data. This is particularly evident for printed documents, where intra-domain data variability is typically low, but inter-domain data variability is high. In that context, current OCR methods do not fully exploit each document's redundancy. We propose an unsupervised method by leveraging the redundancy of character shapes within a document to correct imperfect outputs of a given OCR system and suggest better clustering. To this aim, we introduce an extended Gaussian Mixture Model (GMM) by alternating an Expectation-Maximization (EM) algorithm with an intra-cluster realignment process and normality statistical testing. We demonstrate improvements in documents with various levels of degradation, including recovered Uruguayan military archives and 17th to mid-20th century European newspapers.

The hallmark of Google's Pixel phones has always been the software, whether that's through breakthroughs in computational photography or Google's call screening tech that blocks pesky robocalls. Google today announced the Pixel 10 series, alongside a smartwatch and wireless earbuds, and with them come a slew of new artificial intelligence features ready to wow you. Soniya Jobanputra, director of product management at Google, tells WIRED the Pixel 10 series is all about giving you back time by "taking the mundane and boring out of your life." Let's take a look at exactly how it plans to do that through these software features, some of which may land on older Pixel phones and even other Android phones in the future. That seems to be what Google is implying with its latest addition to the Pixel Camera app: Camera Coach.

This paper introduces a novel approach that leverages Large Language Models (LLMs) and Generative Agents to enhance time series forecasting by reasoning across both text and time series data.

Last year, filmmaker Paul Schrader--the director of Blue Collar, American Gigolo, and First Reformed, and writer of Martin Scorsese's Taxi Driver--issued what seemed like the last word on artificial intelligence in Hollywood filmmaking. A few days after the release of Denis Villeneuve's sci-fi blockbuster Dune: Part Two, Schrader asked his Facebook followers: "Will Dune 3 be made by AI? And, if it is, how will we know?" Schrader is well regarded not only as a director, but one of cinema's top-shelf curmudgeons, quick with a wry burn or baiting shit-post. But his Dune tweet seemed like more than another provocation. It spoke to a mounting feeling among many filmgoers, myself included: that Hollywood had stooped to producing sleek, antiseptic images so devoid of personality that they might as well have been made not by a living, breathing, thinking, feeling artist, but by a computer.