Google DeepMind researchers introduce new benchmark to improve LLM factuality, reduce hallucinations
Based on a new benchmark, Google DeepMind found Gemini 2.0 Flash to be the most factual LLM, with a score of 83.6%.Read More
Read more at VentureBeat
-
Researchers improved AI agent performance on unfamiliar tasks using ‘Dungeons and Dragons’
AgentRefine gives AI agents and models the ability to recognize errors and self-correct to work better for general tasks.VentureBeat - 5h -
Self-invoking code benchmarks help you decide which LLMs to use for your programming tasks
LLMs are good at coding simple functions. But how good are they at calling their own functions to solve complex problems?VentureBeat - 13h -
Alphaville LLC issues new research report on Greenland
A sell-side icebreakerFinancial Times - 2d -
Universities need new ways to make their research pay
Public and private funding are both vital for institutions such as the University of York as international student fees fallFinancial Times - 2d -
Meta proposes new scalable memory layers that improve knowledge, reduce hallucinations
According to Meta, memory layers may be the the answer to LLM hallucinations as they don't require huge compute resources at inference time.VentureBeat - 3d -
New FDA guidance aims to improve accuracy of pulse oximeters for people of color
A Food and Drug Administration (FDA) draft guidance would require manufacturers of pulse oximeters to gather far more clinical data to show the devices accurately work across a range of skin ...The Hill - 4d -
Paxlovid Improved Long Covid Symptoms in Some Patients, Researchers Report
But the report, on the experiences of 13 patients, found that the drug had no benefit for some people and that some who benefited said the improvement didn’t last.The New York Times - 4d -
New York becomes first US city to introduce congestion charging
Plan to ease traffic faced years of challenges and opponents want incoming Trump administration to block itFinancial Times - 5d -
China Introduces New Forex Rules To Tighten Crypto Oversight and Target Illegal Cross-Border Transactions
Yahoo News - Jan. 2
More from VentureBeat
-
King launches Candy Crush Solitaire on mobile devices in February
King announced that it will launch Candy Crush Solitaire on mobile devices on February 6, 2025. It’s a sweet new chapter for Candy Crush, which is an extremely popular mobile game series. And it’s ...VentureBeat - 3h -
OpenAI has begun building out its robotics team
Previously, OpenAI partnered with robotics startup Figure to provide the models powering the latter's humanoid robots.VentureBeat - 4h -
Researchers improved AI agent performance on unfamiliar tasks using ‘Dungeons and Dragons’
AgentRefine gives AI agents and models the ability to recognize errors and self-correct to work better for general tasks.VentureBeat - 5h -
My Arcade reveals Gamestation Retro devices bundled with games
My Arcade revealed its new line of retro consoles, the Gamestation, which come in multiple form factors and launch later this year.VentureBeat - 8h -
Breaking the data bottleneck: Salesforce’s ProVision speeds multimodal AI training with image scene graphs
Salesforce is using structured representation of image semantics to power programs that synthesize instruction datasets for AI training.VentureBeat - 9h
More in Tech
-
Text messaging scammers stole $2M in cryptocurrency from victims, NY attorney general's office says
New York's attorney general says scammers stole millions of dollars in cryptocurrency from people seeking remote work opportunities as part of an elaborate schemeABC News - 1h -
TikTok's Founder Has a Formula for Everything. Can It Crack the Supreme Court?
As the Supreme Court weighs a ban on the video-sharing site, no one has more to lose than Zhang Yiming.The Wall Street Journal - 2h -
Virtua Fighter 5 REVO PC Preorders Are 30% Off
Virtua Fighter will be available on Steam for the first time ever thanks to the new Virtua Fighter 5 REVO edition launching January 27. Fans can snag a discounted preorder for the Steam version at ...GameSpot - 2h -
Candy Crush Solitaire Brings The Classic Card Game To The Massive Mobile Franchise
Candy Crush is getting another feather in its cap. Rather than offering another twist on match-three puzzling, Candy Crush Solitaire will bring the franchise's sickly sweet aesthetic to the card ...GameSpot - 3h -
King launches Candy Crush Solitaire on mobile devices in February
King announced that it will launch Candy Crush Solitaire on mobile devices on February 6, 2025. It’s a sweet new chapter for Candy Crush, which is an extremely popular mobile game series. And it’s ...VentureBeat - 3h