By ResearchBuzz
NEW RESOURCES
National Library of Finland: Käärijä’s Eurovision journey captured in the Finnish Web Archive . “Käärijä came to international attention when he represented Finland in this year’s Eurovision Song Contest, finishing in second place. The National Library of Finland has now collected online materials on the country’s most talked-about phenomenon of the spring 2023.”
Lethbridge News Now: Statistics Canada launches tracking tool as Canada’s population nears 40 million. “Canada is close to reaching 40 million people, and Statistics Canada has launched a new tool to watch that population growth in real time. The Canada population clock keeps an ongoing tally of births, deaths, immigrants, emigrants, non-permanent residents and inter-provincial migrants. The tracker also keeps count of the population of every province and territory, as well as the population change since midnight.”
TWEAKS AND UPDATES
Euronews: ChatGPT and Google Bard adoption remains surprisingly low. “The study, which was conducted in April of this year, included 2 000 people and was focused on AI adoption in addition to people’s willingness to use new AI tools such as ChatGPT and Bard. The results showed that only 19 per cent of the people who took part in the study said that they have used ChatGPT before, while only 9 per cent of the respondents have used the Google Bard chatbot.” This was April… but it’s still lower than I would have thought.
The Verge: Google Chrome’s password manager adds biometric unlocking on desktop. “Google Chrome’s password manager will soon support biometric authentication on PCs and Macs. The feature, which was previously only available on mobile, uses facial recognition or your fingerprint to verify your identity before Chrome automatically fills your passwords.”
Stability.AI: Clipdrop Launches Uncrop: The Ultimate Aspect Ratio Editor. “Today Clipdrop is excited to announce Uncrop, it’s AI-generated ‘outpainting’ tool that allows to change the ratio of any image by creating an expanded background to complement any existing photo or image. Try it for free in the Clipdrop web app – with no need to log in!”
USEFUL STUFF
WIRED: How to Download Your Reddit Data. “Reddit is about to shut off public API access, which means it’s about to get harder to use—and harder to get your data out. Here’s how to grab it now.”
AROUND THE INTERNET WORLD
CBC: Black Loyalist museum was prepared to save artifacts by ‘any means’ during wildfire. “When a massive wildfire started moving north toward the Town of Shelburne last week, Andrea Davis couldn’t help but think back to another fire that had traumatized the community. Davis, the executive director of the Black Loyalist Heritage Society, operates its museum in Birchtown, only a few kilometres southwest of Shelburne.”
SECURITY & LEGAL
Reuters: Social media moderators in Germany seek improved working conditions in battling toxic content. “Hundreds of social media moderators in Germany – who remove harmful content from platforms such as Facebook and TikTok – are calling on lawmakers to improve their working conditions, citing tough targets and mental health issues.”
Deutsche Welle: Ukraine: Cultural heritage sites damaged after dam burst. “Churches, monuments and museums are submerged all over the Kherson region. Archaeological sites dating back to the Scythians — a nomadic people who lived in the region in the 8th century BC — and a Greek settlement from around 400 BC have been damaged or irretrievably destroyed.”
CNN: Radio New Zealand investigates ‘inappropriate editing’ of Ukraine war stories. “New Zealand’s national radio broadcaster has launched an investigation and put a staff member on leave after it said a series of news stories on its website about Russia’s invasion of Ukraine had been edited to present ‘a false account of events.'”
RESEARCH & OPINION
MIT: MIT researchers make language models scalable self-learners. “The scientists used a natural language-based logical inference dataset to create smaller language models that outperformed much larger counterparts.”
Ars Technica: Researchers discover that ChatGPT prefers repeating 25 jokes over and over. “On Wednesday, two German researchers, Sophie Jentzsch and Kristian Kersting, released a paper that examines the ability of OpenAI’s ChatGPT-3.5 to understand and generate humor. In particular, they discovered that ChatGPT’s knowledge of jokes is fairly limited: During a test run, 90 percent of 1,008 generations were the same 25 jokes, leading them to conclude that the responses were likely learned and memorized during the AI model’s training rather than being newly generated.” Good morning, Internet…
Do you like ResearchBuzz? Does it help you out? Please consider supporting it on Patreon. Not interested in commitment? Perhaps you’d buy me an iced tea. Check out Search Gizmos when you have a minute.
June 16, 2023 at 05:31PM
via ResearchBuzz https://ift.tt/hvkSLN8
No comments:
Post a Comment