Whenever I fantasise about a couple of hours of uninterrupted relaxation during the chilly winter months, my mind immediately conjures up images of curling up on the sofa with a deliciously good book. And when summer eventually comes around, just swap the location to a sun lounger in the back garden (or somewhere more exotic). So with 2026 nearly upon us, join me for an eclectic taste of a few literary delights worth feasting upon over the next 12 months. It's the final instalment of Oseman's hit graphic novel series which has followed the lives of Nick and Charlie, two teenage boys who fall for each other at school. Along with their friends, we've followed all the ups and downs of their relationship as they navigated family drama, homophobia and mental health issues, alongside the joy of first love.

Recent advances in text-to-image (T2I) generation have achieved impressive results, yet existing models often struggle with simple or underspecified prompts, leading to suboptimal image-text alignment, aesthetics, and quality. We propose a prompt rewriting framework that leverages large language models (LLMs) to refine user inputs before feeding them into T2I backbones. Our approach introduces a carefully designed reward system and an iterative direct preference optimization (DPO) training pipeline, enabling the rewriter to enhance prompts without requiring supervised fine-tuning data. We evaluate our method across diverse T2I models and benchmarks. Results show that our prompt rewriter consistently improves image-text alignment, visual quality, and aesthetics, outperforming strong baselines. Furthermore, we demonstrate strong transferability by showing that a prompt rewriter trained on one T2I backbone generalizes effectively to others without needing to be retrained. We also systematically study scalability, evaluating how performance gains scale with the capacity of the large LLM used as the rewriter. These findings highlight that prompt rewriting is an effective, scalable, and practical model-agnostic strategy for improving T2I systems. We plan to release the code and trained prompt rewriters soon.

Atlantic writer Faith Hill claimed that women have developed a distrust of dating apps due to young men becoming more conservative during an appearance on CNN's "The Assignment with Audie Cornish" on Thursday. Young men's shift to the political right has complicated the dating world and led to distrust by women of dating apps, according to The Atlantic writer Faith Hill, who appeared on CNN on Thursday. Hill argued that women's growing distrust of dating apps stems from men -- young men in particular -- becoming more conservative while young women are becoming more progressive, leading to the sexes "growing further apart in a lot of ways." "You see that young men are moving further to the right. And I think for a lot of women in particular, it can just sort of feel like, 'This is not a time where I trust men -- I feel respected by men. I don't necessarily want to go out and meet strangers who are men,'" Hill said.

Although chain-of-thought reasoning and reinforcement learning (RL) have driven breakthroughs in NLP, their integration into generative vision models remains underexplored. We introduce ReasonGen-R1, a two-stage framework that first imbues an autoregressive image generator with explicit text-based "thinking" skills via supervised fine-tuning on a newly generated reasoning dataset of written rationales, and then refines its outputs using Group Relative Policy Optimization. To enable the model to reason through text before generating images, We automatically generate and release a corpus of model crafted rationales paired with visual prompts, enabling controlled planning of object layouts, styles, and scene compositions. Our GRPO algorithm uses reward signals from a pretrained vision language model to assess overall visual quality, optimizing the policy in each update. Evaluations on GenEval, DPG, and the T2I benchmark demonstrate that ReasonGen-R1 consistently outperforms strong baselines and prior state-of-the-art models. More: aka.ms/reasongen.

When you're not watching Barkitecture or the Weird Al movie on your Roku, the device turns on its familiar scrolling, purple city-scape. Now you don't just have to settle for a Flintstones-esque background of neoclassical buildings and the occasional billboard for PlutoTV when your Roku goes into rest mode. The TV viewing platform is introducing a new screensaver feature called Backdrops. The new viewing feature turns your TV or viewing device into an Amazon Echo Show 8 Photos Edition, except Roku hasn't taken away the only feature that gave it its name. Backdrops will show pictures of classic works of art by masters like Claude Monet and beautiful photos of landscapes that you can choose to display on your Roku device.

Every day, there seems to be new generative AI news, and while it can often be serious and quite technical, this time around it's just plain fun. Instagram has launched a new generative AI-powered tool called backdrop that lets you create a new image in the, yes, background of your Story. Meta's generative AI lead, Ahmad Al-Dahle, announced the feature on Threads alongside a video tutorial. Instagram's backdrop tool appears once you upload or capture content for your Story. It sits alongside existing icons at the top of your screen, like text and music, represented by an image of a person with a rectangular frame behind them.

Meta is introducing a virtual assistant today to compete with OpenAI's ChatGPT that can serve up answers to questions from Microsoft's Bing search engine and generate images from text commands. Meta AI, as the assistant is called, is powered by the company's large language model Llama 2. As well as chatting it can generate images, using a new image generator named Emu that Meta trained on 1.1 billion pairs of photos and text, including photos and captions shared on Facebook or Instagram. The new assistant will be available today for a limited group of US users on Facebook Messenger, Instagram, and WhatsApp. It can be added to group chats and help out with tasks such as making travel plans. The assistant will also be available via voice using new smart glasses Meta will release next month for US users.

Meta is starting to make good on its promise to bring generative AI to all of its products. At the company's Connect event, it revealed new AI image editing and sticker-creation features for Instagram. A tool called "restyle" is a bit like a supercharged generative AI filter. It allows users to remix their existing photos into different looks. "Think of typing a descriptor like'watercolor' or a more detailed prompt like'collage from magazines and newspapers, torn edges' to describe the new look and feel of the image you want to create," the company explained.

When Y2K seemed like the world's most pressing technological concern, the Mexican-Canadian artist Rafael Lozano-Hemmer was using a dictionary and a set of grammatical rules to teach a computer how to write questions. The program he built can make enquiries in Spanish, English, German and French, in 4.7tn possible combinations. When the artwork showed at the San Francisco Museum of Modern Art last year, it still had 271,000 years of new questions to ask. Which is to say, Lozano-Hemmer has been working with generative technology long enough to have learned a powerful lesson: "There is no such thing as a neutral algorithm." This lesson was reiterated to the Bafta-winning media artist in a spectacular, humiliating fashion at Miami Art Basel a little over a decade ago.

The manufacturing industry is facing a plethora of external challenges: supply chain disruptions, labour shortages and vast globalisation are amongst those at the top of many lists. Against the backdrop of vast international competition, these external factors have undoubtedly created heightened pressures for manufacturers to outperform their counterparts. Could AI and AR be the solution? Artificial intelligence and the Internet of Things (IoT) are fundamentally redesigning the manufacturing landscape by equipping manufacturers to reassess their current processes and build long-term competitive advantages that enables them to adapt to current and future challenges. When considering potential areas for operational improvement and productive efficiency gains, manufacturers shouldn't underestimate the impact of their machinery and internal equipment – particularly with regards to maintenance and malfunctions. Not only does infrequent routine maintenance and the inability to put operations back online cause an immediate and frustrating end to productivity, but it also creates time-consuming and costly delays that create the potential for competitors to bypass the manufacturer in question.