Viktor Antonov, best known for his work as art lead on Half-Life 2 and Dishonored, has reportedly died at age 52. Half-Life writer Marc Laidlaw broke the news in an Instagram Story, and other colleagues have since taken to social media to pay tribute as well. "I didn't want to say much till I felt it was confirmed, but I learned today that Viktor Antonov, our visionary art lead on HL2, has died," Laidla wrote in the now-expired post, which was reshared by LambdaGeneration on Saturday night. Antonov got his start in video games working on Redneck Rampage, and in addition to serving as art director for Half-Life 2 and Dishonored, he went on to consult on titles including Doom (2016) and Fallout 4. The Bulgarian artist just recently appeared in a documentary celebrating the 20th anniversary of Half-life 2 this past November. I wish I told you how much admiration I had for you but we get caught in our lives until a surprise like this hits us," Raphael Colantonio, founder of Arkane Studios and Wolfeye Studios, wrote on Bluesky. "You were instrumental to the success of Arkane Studios and an inspiration to many of us, also a friend with whom I have many fond memories." In another post, game designer Harvey Smith added, "All this about his impact and talent is true, but I will also always remember how much he made me laugh, with his dry, devastating wit.

Recent advancements in text-to-image generation have revolutionized numerous fields, including art and cinema, by automating the generation of high-quality, context-aware images and video. However, the utility of these technologies is often limited by the inadequacy of text prompts in guiding the generator to produce artistically coherent and subject-relevant images. In this paper, We describe the techniques that can be used to make Large Language Models (LLMs) act as Art Directors that enhance image and video generation. We describe our unified system for this called "LaDi". We explore how LaDi integrates multiple techniques for augmenting the capabilities of text-to-image generators (T2Is) and text-to-video generators (T2Vs), with a focus on constrained decoding, intelligent prompting, fine-tuning, and retrieval. LaDi and these techniques are being used today in apps and platforms developed by Plai Labs.

Humans can reason compositionally when presented with new tasks. Previous research shows that appropriate prompting techniques enable large language models (LLMs) to solve artificial compositional generalization tasks such as SCAN. In this work, we identify additional challenges in more realistic semantic parsing tasks with larger vocabulary and refine these prompting techniques to address them. Our best method is based on least-to-most prompting: it decomposes the problem using prompting-based syntactic parsing, then uses this decomposition to select appropriate exemplars and to sequentially generate the semantic parse. This method allows us to set a new state of the art for CFQ while requiring only 1% of the training data used by traditional approaches. Due to the general nature of our approach, we expect similar efforts will lead to new results in other tasks and domains, especially for knowledge-intensive applications.

"Can a robot write a symphony? Can a robot turn a canvas into a beautiful masterpiece?" So asks Will Smith in the 2004 movie iRobot. Of course, the answer back then was a resounding'no', however, 15 years later it's not so simple. An AI rapper has already been signed to a record label.

It's August 2022, and by now you've no doubt read (or more likely seen) something about AI art by now. Whether it's random jokes made for Twitter or paintings that look like they were made by actual human beings, artificial intelligence's ability to create art has exploded onto the scene over the last few months, and while this has been great news for shitposts and fans of tech, it has also raised a number of important questions and concerns. If you haven't read or seen anything about the subject, AI art--or at least as it exists in the state we know it today--is, as Ahmed Elgammal writing in American Scientist so neatly puts it, made when "artists write algorithms not to follow a set of rules, but to'learn' a specific aesthetic by analyzing thousands of images. The algorithm then tries to generate new images in adherence to the aesthetics it has learned." Currently there are a handful of prominent platforms that people are using, with three of the most popular being Midjourney, Dall-E and Stable Diffusion.

If you ask multiple technical artists what they do, the chances are you'll get very different responses. Although there are generalists, it is more often an umbrella that encompasses a number of specialist disciplines such as environment, shaders, VFX, and pipeline, to name just a few, so you could be involved more with the front-end or the back-end. Naturally, this also varies hugely from studio to studio. However, if we're to explain the essence of the role, then Jodie Azhar, technical art director at Silver Rain Games says: "Technical artists are problem solvers." "Quite often they will be creating tools to solve these problems, for instance creating a procedural system to save time for an environment artist from doing very repetitive tasks. It also involves being the bridge between the art and programming sides of game development. "They can act like translators between artists and programmers," she adds. "They need to understand artists' processes and how they want something to ...

The next few Tuesdays, The Verge's flagship podcast The Vergecast is showcasing a miniseries dedicated to the use of artificial intelligence in industries that are often overlooked, hosted by Verge senior reporter Ashley Carman. This week, the series focuses on AI for the video world. More specifically, we're looking at how AI is being used as a tool to help people streamline the process of creating video content. Yes, this might mean software taking on a bigger role in the very human act of creativity, but what if instead of replacing us, machine learning tools could be used to assist our work? That's what Scott Prevost, VP of Adobe Sensei -- Adobe's machine learning platform -- envisions for Adobe's AI products. "Sensei was founded on this firm belief that we have that AI is going to democratize and amplify human creativity, but not replace it," Prevost says.

Nikolay Ironov had been working as a graphic designer for more than a year before he revealed his secret. Lebedev Studio -- Russia's largest design company -- Ironov had already worked on more than 20 commercial projects, creating everything from beer bottle labels to startup logos. But Ironov was not the person he claimed to be. In fact, the designer was not a person at all. Lebedev Studio revealed the truth to its clients: their logos had been created by an AI system.

A brand manager needs an advertisement. So, the brand manager sends a brief to the senior art director (in-house or at an agency) and asks for something amazing to be created. On or before the deadline, the brand manager and the art director meet to review the work. The brand manager is presented with three approaches, and after a number of meetings, a number of revisions and revelations, they agree on a final product. This is a process that has repeated itself for more than a century, and AI is not going to stop it.

State-of-the-art machine learning methods exhibit limited compositional generalization. At the same time, there is a lack of realistic benchmarks that comprehensively measure this ability, which makes it challenging to find and evaluate improvements. We introduce a novel method to systematically construct such benchmarks by maximizing compound divergence while guaranteeing a small atom divergence between train and test sets, and we quantitatively compare this method to other approaches for creating compositional generalization benchmarks. We present a large and realistic natural language question answering dataset that is constructed according to this method, and we use it to analyze the compositional generalization ability of three machine learning architectures. We find that they fail to generalize compositionally and that there is a surprisingly strong negative correlation between compound divergence and accuracy. We also demonstrate how our method can be used to create new compositionality benchmarks on top of the existing SCAN dataset, which confirms these findings.