Meta新AI模型“Maverick”涉嫌优化“作弊”，引发业内质疑
Meta近日发布的新旗舰AI模型“Maverick”在LM Arena排行榜上名列第二，但研究人员指出，该测试版本与开发者可用的公开版本存在明显差异。
Meta承认，参与测试的是“针对对话优化”的实验性版本。研究者发现，LM Arena上的Maverick回答更冗长，使用大量表情符号，而公开版本则表现平平。此举引发业内对AI公司是否应为排行榜优化模型的争议。过度定制测试版本可能误导开发者，破坏评测公正性。Meta与LM Arena官方目前尚未回应相关质疑。
TechCrunch
📮投稿 ☘️频道 🌸聊天
BY 科技圈🎗在花频道📮
The campaign, which security firm Check Point has named Rampant Kitten, comprises two main components, one for Windows and the other for Android. Rampant Kitten’s objective is to steal Telegram messages, passwords, and two-factor authentication codes sent by SMS and then also take screenshots and record sounds within earshot of an infected phone, the researchers said in a post published on Friday.
The global forecast for the Asian markets is murky following recent volatility, with crude oil prices providing support in what has been an otherwise tough month. The European markets were down and the U.S. bourses were mixed and flat and the Asian markets figure to split the difference.The TSE finished modestly lower on Friday following losses from the financial shares and property stocks.For the day, the index sank 15.09 points or 0.49 percent to finish at 3,061.35 after trading between 3,057.84 and 3,089.78. Volume was 1.39 billion shares worth 1.30 billion Singapore dollars. There were 285 decliners and 184 gainers.
在花软件新闻投稿频道 from tw