Genomic language models (gLMs) hold promise for generating novel, functional DNA sequences for synthetic biology. However, realizing this potential requires models to go beyond evolutionary plausibility and understand how DNA sequence encodes gene expression and regulation. We introduce a benchmark called Nullsettes, which assesses how well models can predict in silico loss-of-function (LOF) mutations, in synthetic expression cassettes with little evolutionary precedent. Testing 12 state-of-the-art gLMs, we find that most fail to consistently detect these strong LOF mutations. All models show a sharp drop in predictive accuracy as the likelihood assigned to the original (nonmutant) sequence decreases, suggesting that gLMs rely heavily on pattern-matching to their evolutionary prior rather than on any mechanistic understanding of gene expression. Our findings highlight fundamental limitations in how gLMs generalize to engineered, non-natural sequences, and underscore the need for benchmarks and modeling strategies that prioritize functional understanding.

Do you, like me, have paper documents that have long since been scanned and processed, records or music cassettes that you would like to listen to on your mobile phone, and photo prints that are planned for a digital photo book? Then you will appreciate the two-step instructions in this article, with which you can convert analog media to digital and then process them further. Important insurance papers, contracts, invoices, or simply the page-long letter from your favorite aunt -- there are many paper documents that you want to scan in order to preserve them. If it's even a text that you want to search and edit, you can run OCR software over it after scanning, which recognizes the text so that you can search it and, if necessary, edit it with a standard word processor. With the freeware Not Another PDF Scanner 2 (Naps 2), you have plenty of options for editing and saving the scan after scanning a document.

Following a lot of hype--and a quarter-century of work--"Now and Then," presumably the last song to feature all four original Beatles, is here. The track dropped Thursday and the music video, directed by Peter Jackson, hit YouTube on Friday. Sweet and haunting, it's full of piano and strings, and it wouldn't have been possible without the machine learning technology Jackson used on the docuseries Get Back. How the AI technology became the thing that saved the song is a bit of a journey. Years after John Lennon died in 1980, his wife, the musician and multimedia artist Yoko Ono, told his bandmate Paul McCartney that she had a demo tape Lennon had recorded at their apartment in the Dakota in New York City.

Artificial intelligence has helped create a final Beatles song set to be released this year, its member Paul McCartney has said. In an interview released on Tuesday by the BBC, McCartney said the technology was used to "extricate" John Lennon's voice from an old demo which was used to complete the song. "We just finished it up, and it'll be released this year," he said. McCartney, 84, said the song was made with the help of film director Peter Jackson, using the same AI technology employed for the Beatles documentary Get Back. During the making of that film, Jackson and his team were able to separate the voices from the instruments.

It's almost unthinkable now, but from the 1970s until the early 1980s, vinyl records were explored as a means of storing computer data – including video games. Some magazines of the time tucked code-packed flexi disc inserts into their pages: paper-thin plastic records that could be fed into home computers from an ordinary turntable, magically manifesting a game on screen. Long before Travis Scott was attracting 12 million players to a gig hosted in Fortnite, there was a coming together of a British game developer, a magazine and a pop act that marked the beginning of the intersection between the music and games industries. The Thompson Twins Adventure Game came cover-mounted on a 1984 issue of the beloved magazine Computer & Video Games, the first UK magazine devoted to games. Almost everyone involved in the project – a promotional item linked to the release of the single Doctor Doctor – admits the game was imperfect. It was a weird text adventure garnished with incidental visuals, in which the members of the Thompson Twins had to locate the ingredients of a potion to be made by the song's eponymous medic.

A byte of data has been stored in a number of different ways through the years as newer, better, and faster storage media are introduced. A byte is a unit of digital information that most commonly refers to eight bits. A bit is a unit of information that can be expressed as 0 or 1, representing a logical state. Let's take a brief walk down memory lane to learn about the origins of bits and bytes. Going back in time to Babbage's Analytical Engine, you can see that a bit was stored as the position of a mechanical gear or lever. In the case of paper cards, a bit was stored as the presence or absence of a hole in the card at a specific place. For magnetic storage devices, such as tapes and disks, a bit is represented by the polarity of a certain area of the magnetic film. In modern DRAM (dynamic random-access memory), a bit is often represented as two levels of electrical charge stored in a capacitor, a device that stores electrical energy in an electric field. In June 1956, Werner Buchholz coined the word byte to refer to a group of bits used to encode a single character of text. Let's address character encoding, starting with ASCII (American Standard Code for Information Interchange). ASCII was based on the English alphabet; therefore, every letter, digit, and symbol (a-z, A-Z, 0-9,, -, /, ",!, among others) were represented as a seven-bit integer between 32 and 127. To support other languages, Unicode extended ASCII so that each character is represented as a code-point, or character; for example, a lowercase j is U 006A, where U stands for Unicode followed by a hexadecimal number. UTF-8 is the standard for representing characters as eight bits, allowing every code-point from 0 to 127 to be stored in a single byte. This is fine for English characters, but other languages often have characters that are expressed as two or more bytes.

In a recent conference call to discuss its latest quarterly financial results, the games publisher Take Two provided some astonishing statistics about Grand Theft Auto V. According to the company's CEO, Strauss Zelnick, the open-world gangster adventure, originally released in 2013, has now sold more than 75m copies. Not only that, but NPD Group sales data shows it was the sixth best-selling game across all formats in 2016. If you look at the current UK games chart, GTA V is at number two, beaten only by Resident Evil 7, released last month. There are some obvious reasons.

Think of the gear you can't live without: The smartphone you constantly check. The camera that goes with you on every vacation. The TV that serves as a portal to binge-watching and -gaming. Each owes its influence to one model that changed the course of technology for good. Some of these, like Sony's Walkman, were the first of their kind. Others, such as the iPod, propelled an existing idea into the mainstream. Some were unsuccessful commercially, but influential nonetheless. And a few represent exciting but unproven new concepts (looking at you Oculus Rift). Rather than rank technologies--writing, electricity, and so on--we chose to rank gadgets, the devices by with consumers let the future creep into their present. The list--which is ordered by influence--was assembled and deliberated on at (extreme) length by TIME's technology and business editors, writers and reporters.

The hyper-modular printer control project at PARC has proven that a tightly integrated model-based planning and control framework can effectively control a complex physical system. Recently, we have successfully applied this framework to another application: planning for the Material Control System (MCS) of Liquid Crystal Display (LCD) manufacturing plant in a joint project between the Embedded Reasoning Area at PARC and the Products Development Center at the IHI Corporation. The model-based planner created at PARC was able to successfully solve a diverse set of test scenarios provided by IHI, including those that were deemed very difficult by the IHI experts. The short projecttime (2 months) proved that model-based planning is a flexible framework that can adapt quickly to novel applications. In this paper, we will introduce this complex domain and describe the adaptation process of the Plantrol online planner. The main contributions are: (1) introducing a successful application of general-purpose planning; (2) outline the timeline-based online temporal planner; and (3) description of a complex warehouse management problem that can serve as an attractive benchmark domain for planning.