Mistral, founded by the nephew of Laurent Alexandre, who has received hundreds of millions of euros in public contracts and constant support from the Élysée, has given up on producing generalist LLMs. In doing so, it leaves France without a player in this crucial field. If we (translated)
LLM-speak is very invasive. Like any other style of language or slang. When there are many users of some constructions and we encounter them frequently, we tend to imitate them. I expect the future of human language to be symbiotic with AI language use. #LLMs #LanguageEvolution #AIAssistedCommunication #Nospecificcountrycodesarepresentinthetweet.
RECORDEMOS: Con este gol que se comió Martinez Quarta (enfrente mío), River se CLASIFICABA a la próxima ronda del MUNDIAL de CLUBES. El rival? El PEOR DORTMUND del siglo. https://t.co/llmAKqRNf7 #RiverPlate #MundialDeClubes #Dortmund
Santos encaminha contratação de Arthur e corre para derrubar transferban. (com a @llmarcello_). Saiba mais: https://t.co/wSNYM9fPkA #SantosFC #Arthur #TransferBan
bez cienia żenady pada tu wspaniała(ale niestety przerwana przez Jankowskiego) myśl z ust pani profesor, że co drugi profesor na świecie wykorzystuje AI do pracy XDDDDDDDD innego końca świata nie będzie, witaj w rzeczywistości „nauki” pod tezę stworzonej przez halucynacje LLM-ów #AI #nauka #technologia
The study shows that large language models (LLMs) make conclusions about people's character traits based on their faces, which aligns with problematic human judgments, while most respondents prefer one of the generated faces, illustrating the persistent biases that are transferred to technology. (translated)
Ibland undrar jag om Strandhälls mo är att förleda uppmärksamheten och dölja något betydligt värre. En sorts välkalibrerad LLM med syftet att provocera på lägsta möjliga vis. #politik #medvetenhet #samhälle
"LLM answers sound right—the machine provides confident prose—but to get the correct answer, you would need to know to press it in ways that are not obvious to most people." Exactly. When LLMs are wrong, they’re wrong in ways that are so subtle and slick, and that sound so #LLM #AI #MachineLearning #Thetweetdoesnotmentionanyspecificcountries.Therefore #therearenocountrycodestoreturn.
Nowadays, asking AI is the best way to win an argument. But in most cases the trick is how the question is posed: frame it one way, it agrees, frame it another, it doesn't. We handed LLMs the role of judge but people are learning to ask questions based on what answers they want. https://t.co/sYzHu21bW5 #AI #Argumentation #LLMs #Nospecificcountriesarementionedinthetweet.
AI způsobila, že na sociálních sítích už prostě ignoruju delší texty. Tak nějak předpokládám, že je napsal LLM. #AI #sociálníSítě #LLM
The article discusses the significant advances of artificial intelligence systems in mathematics, highlighting both recent successes in solving historical conjectures and a persistent mistrust towards such technologies, due to their surprising similarity to human behavior. (translated)
Very interesting study. From the response of an LLM, the original prompt can be determined through backward token prediction. This can reveal many things, from trade secrets to medical data, that one thought were protected. The study needs to be reproduced. (translated)
Worth seeing, a good historical overview of the emergence of LLM from the experience of a pioneer, explaining important technical breakthroughs. (translated)
Helt fint - men det skal altså holdes op imod at folk i forvejen ikke ved hvad en kvinde er, mener at Kommunisme er den rigtige vej og tror at verden går under grundet CO2. Folk er idioter. Kunstig intelligens kan umulig gøre dem dummere end dette. https://arstechnica.com/ai/2026/04/research-finds-ai-users-scarily-willing-to-surrender-their-cognition-to-llms/
One of the most telling results in LLMs history is that models that didn't do any training to produce a chain of thoughts would perform better if asked to think about problems. Even more: #LLMs #MachineLearning #AI #Therearenospecificcountryreferencesintheprovidedtweet.Therefore #therearenocountrycodestoreturn.
I'll regret cramming it into a tweet, but among folks who see LLMs as transformative, I'm baffled that so few try to carve out a new path for their craft. It's either "we're cooked and nothing makes sense anymore" or "I'll get ahead because I prompt better than anyone else". #LLMs #transformation #creativity #Nospecificcountrycodescanbederivedfromthetweetprovided.
Reddit fungeert als een belangrijke bron voor kunstmatige intelligentie, zoals ChatGPT, maar dit trekt ook spam en desinformatie aan, waardoor het platform kwetsbaar is, zoals blijkt uit recent onderzoek van Cornell University. #Reddit #KunstmatigeIntelligentie #LLMPoisoning
For decades, getting featured in a music magazine like Tunitemusic was primarily about building human connection. A well-written, deeply descriptive album review or an insightful artist feature allowed independent musicians, contemporary composers, and instrumental artists to share their narrative, establish cultural credibility, and reach a dedicated audience of music lovers. While the human element of music journalism remains entirely irreplaceable, a quiet revolution has taken place behind the digital curtain. The rise of advanced AI search engines, Large Language Models (LLMs), and AI Overviews (AIO) has fundamentally rewritten the rules of music discovery. Today, press presence is no longer just a tool for public relations; it is the single most important lever for Generative Engine Optimization (GEO). If you want artificial intelligence to recognize you, recommend your compositions, and talk about your music, you need professional reviews written about your work. #MusicPromotion #MusicJournalism #IndependentMusician #MusicReviews #ComposerLife #MusicPR #Neoclassical #AmbientPiano #MusicDiscovery #IndependentArtist #GenerativeAI #MusicMarketing #ContemporaryClassical #Tunitemusic #ArtistsOfTomorrow https://tunitemusic.com/post/ai-search-is-changing-how-music-fans-discover-you/
Academic writing with LLM has unexpected benefits, like getting explanations of how to position an interdisciplinary paper so the two main disciplines it deals with would like it. I would not have seen it. #AcademicWriting #Interdisciplinary #LLM #Nospecificcountrycodes.
