Archive: August 2020

The natural language processing (NLP) field has achieved remarkable advancements spurred by the power of transformer-based pretrained large language models such as Google’s BERT and OpenAI’s GPT. Transformer architectures have recently also attained state-of-the-art performance on various computer vision tasks. Efforts to extend the BERT-style masked image modelling pretraining approach (where a portion of the image is masked and the model learns to recove

Credit: CC0 Public Domain Bacteria are the most diverse organisms on Earth, with a number of species that's difficult to quantify. They're also incredibly old Bacteria consist of a single cell. They do not have bones and are not like big animals that leave clear signs in the geological record, which thankful paleontologists can study many millions of years later. This has made it very hard for scientists to establish a timeline of their early evolution. Bu

Contemporary text-to-speech (TTS) systems are achieving increasingly impressive humanlike voice generation performance in terms of intelligibility and naturalness. A lack of diverse speech data however makes it challenging to faithfully capture the many different speaker identities, prosodies and styles that can present in human speech. In the new paper NaturalSpeech 2: Latent Diffusion Models are Natural and Zero-Shot Speech and Singing Synthesizers, a Mi

Razer’s next big black-and-green gamer chair is designed to make you hear and feel every in-game gunshot or the spine-rattling rumble of a speeding motorcycle. The Razer Soma Pro takes the same basic ergonomic design of the Soma Chroma from earlier this year but adds in back and bottom haptics alongside speakers built into the neck rest. There are two new chairs filling out Razer’s Soma line, one called the Soma Sensa HD Haptics and the other a top-end thr

UK gears up for fight against Russia’s disinformation machine The UK government will create a new body to track and disrupt disinformation campaigns run by hostile states, Prime Minister Andy Burnham announced at the United Nations General Assembly in New York. In his first address to the Assembly on 22 September, Burnham told world leaders he was tasking UK security chiefs to begin work on the National Centre for Information Defence, which will operate “t

Stop watching what AI agents say and start watching what they do In this interview with Help Net Security, Ariel Assaraf, CEO of Coralogix, explains why a system prompt can describe a boundary for an AI agent but cannot enforce one. Assaraf covers how his team builds AI agent guardrails at the execution layer, limits context without expanding authority, and tunes controls by risk so agents stay useful during incidents. He describes how he tracks the conseq

Credit: AgriEngineering (2025). DOI: 10.3390/agriengineering7030081 While strawberry production runs from November through April in Florida, digital twin technology lets scientists simulate the growth of the fruit year-round, allowing research to proceed year-round. Digital twins are virtual replicas of objects, systems or processes that can predict system behavior as they interact in a simulated environment. Dana Choi and her team of University of Florida

GitHub adds AI to catch passwords before a code push GitHub has announced an AI detector, developed with Microsoft Applied Sciences, to help prevent developers from uploading passwords and other credentials to code repositories. The ModernBERT-based classifier will expand GitHub’s push protection, which checks code for secrets and can block a push before a credential enters repository history. How the detector works Existing checks recognize many credentia

Music streaming services—like virtually every other corner of the internet—are awash in AI slop, thanks to the rise of cheaply accessible AI tools and a regulatory void. Now, a coalition of organizations representing musicians and record labels is trying to make it easier for listeners to at least know when a song has been generated by an LLM. Led by the Recording Industry Association of America (RIAA) and the International Federation of the Phonographic I

OpenAI CEO Sam Altman said in a Reddit forum early this year that the artificial intelligence star was on 'the wrong side of history' when it came to being open about the workings of its technology. OpenAI on Tuesday released two new artificial intelligence (AI) models that can be downloaded for free and altered by users, to challenge similar offerings by US and Chinese competition. The release of gpt-oss-120b and gpt-oss-20b "open-weight language models"

Artificial intelligence labs are after computing power. Kalshi doesn’t have any to offer them but they’ve got something else that might be of some use, according to Bloomberg: a tool that plots the predicted future price of computing power. So, that’s something! There really is some value for companies to know where the cost of compute is going—especially since currently, it’s mostly just going up. Having a sense of just how much compute is going to cost a

Intel, Microsoft and many homegrown startups also conduct a sizable share of their R&D in Israel. But owing to Nvidia’s stronghold on AI given the centrality of its GPUs for training LLMs and using them, many in the country are wondering how much of a reshuffle will result from its leadership crisis. As with everything at Nvidia, Huang drove the company to acquire Israeli chipmaker Mellanox for $6.9 billion in 2019 and use the company as the base for its l

In recent times, video generative models have emerged as a focal point of attention, unlocking a realm of fresh creative opportunities. Despite this, the velocity of these models remains a significant obstacle to their broader adoption. State-of-the-art generative models, while impressive in their capabilities, are hampered by sluggishness and computational demands due to their iterative diffusion processes. To address this issue, in a new paper AnimateDif