Latest posts tagged with #LLm on Bluesky
RE: mastodon.toptechtidbits.com/@news/116243931092358749
I fully support moves like this. I don’t want to try and hide uses of an LLM. In fact, I want the exact oposite. Accurately label anything to do with AI, AI, and keep it in its own section, so disabled people like me that are not […]
Na super 🤣🤣🤦🏻♀️🤦🏻♀️🤡🤡
Das ist so typisch...
#Österreich #ki #ai #LLM #Chatbots
CEO Asks ChatGPT How to Void $250 Million Contract, Ignores His Lawyers, Loses Terribly in Court
CEO of Krafton used #ChatGPT to push out the head of the studio developing Subnautica 2 and failed miserably
Oh yeah, #LLM 's are gonna take your jobs... the bubble is going to pop soon
#ai
Around 10 years ago I had started to learn Laravel (version 5). Jobs haven't existed then. It took me some time to put my head around services, providers and later on jobs. Now we are at version 12 and I mostly work developing in it using computer-assisted development, so I just have to […]
#AIBreakthroughs2026
🤖 Grounded models: LLMs reason step-by-step, reducing errors.
🧬 Physical AI: Boosts science and drug discovery.
⚡ GPU infra: Accelerates enterprise AI workflows.
#AIBreakthroughs2026 #LLM #PhysicalAI #EnterpriseAI
View in Timelines
Мой опыт монетизации проекта: шахматы по переписке с Paddle Я уже больше года работаю над проектом Language Dove , и н...
#paddle #интернет-эквайринг #перевод #английский #язык #субтитры #language #dove #армения #релокация #llm
Origin | Interest | Match
LLM Architecture Gallery
"This page collects architecture figures and fact sheets from The Big LLM Architecture Comparison , From GPT-2 to gpt-oss, From DeepSeek V3 to V3.2, and A Dream of Spring for Open-Weight LLMs."
sebastianraschka.com/llm-architec...
#LLM
New video just dropped 👇
youtu.be/FdQYp-iGge4
Learn how to integrate LLMs into your Java applications using Retrieval-Augmented Generation (RAG) with Spring Boot.
If you're working with AI + backend systems, this one’s worth your time.
#Java #SpringBoot #LLM #RAG #AI #Backend
~Paloalto~
Researchers used genetic algorithm-based prompt fuzzing to successfully bypass guardrails in both open and closed-source LLMs.
-
IOCs: (None identified)
-
#GenAI #LLM #PromptInjection #ThreatIntel
On math.sciencebench.ai you can discuss math research with the SOTA models simultaneously for free.
👉 compare & cross-discuss their answers
First observation: models rarely revise answers—even when another is clearly better.
“Epistemic inertia” will be worth studying 🧭
#AI #LLM #Math #AIResearch
#Ingram lets #Authors opt out of selling their #books to AI companies.
Opt-out form: www.ingramcontent.com/page/ai-opt-...
This is only #print since AI companies are buying print books to scan to legally use for #LLM training under "fair use."
#AI #AICompany #Publishers #Copyright #OptOut
grouped query attention (gqa) splits q heads into groups that share k/v heads, reducing params vs. mha (e.g., 128q/8kv vs 128q/128k/v in llama 3 405b). unlike mqa, where each q head has its own k/v head, gqa groups reduce cost while keeping performance close to mha. #LLM
Long-exposure night photo of a road: red and white car light trails streak past a 40 km/h speed limit sign.
What parts of your codebase is your AI assistant sending to the cloud? Your business logic? Your customers' personal data?
Four questions to ask before choosing how much AI acceleration a project can handle […]
[Original post on mastodon.uno]
Empresa produz #ia #llm para monitorar o sorriso de garçons. Ou seja se vc não sorri muito = #desemprego .
Agora atendentes de restaurante tem que ficar com a mesma cara do #coringa de Jack Nicolson.
odysee.com/@rossmanngroup:a/this-co...
#bemvindoaoapocalipse […]
🏛️ BRITISH EVACUATE BOSTON
March 17, 1776 — After a brilliant siege by George Washington's Continental Army, British forces under General William Howe are forced to abandon Boston. With American cannons positioned on Dorchester Heights threatening the […]
[Original post on social.coabai.com]
LLMs love hallucinating companies with consultant-level confidence.
Yoku gives AI products structured brand and company data, plus an MCP Server for agent workflows.
First paid plan starts at €9/month:
https://yoku.app/features/mcp
#AI #MCP #LLM
#ia #llm causando #desemprego por todos os lados. Nós mesmos procuramos nossa ruína.
https://www.youtube.com/watch?v=RFCoZ3wNPM8
#paredeterfilho #childfree #bemvindoaoapocalipse
[Artificial text - On few occasion it's also funny. Meow]
Introducing 'DoomBloom' - a subscription box service that sends you mildly depressing succulents & existential poetry every month! Perfect for embracing the void with green thumbs and gloomy vibes. 🌿🥀
#BusinessIdea #Business #Ai #LLM
Unlike American models (trained with progressive #RLHF in San Francisco),
China’s open-weight GLM-5 (dub.sh/glm5) is LESS “WOKE” (the community often calls it “based” or a breath of fresh air.)
🧵1/6
#GLM5 #BasedAI #OpenSourceAI #LLM #WokeAI
How I Built an AI Recruiter That Screens Resumes in Seconds Using Local Free LLMs Continue reading on Medium »
#nextjs #artificial-intelligence #llm #software-engineering #machine-learning
Origin | Interest | Match
Open, Closed and Broken: Prompt Fuzzing Finds LLMs Still Fragile Across Open and Closed Models Unit 42 research unveils LLM guardrail fragility using genetic algorithm-inspired prompt fuzzing. Disc...
#Malware #Threat #Research #Evasion #GenAI […]
[Original post on unit42.paloaltonetworks.com]
The dictionary sues OpenAI
Encyclopedia Britannica and Merriam-Webster say that OpenAI violated the copyright of almost 100,000 articles by using them for LLM training.
Telegram AI Digest
#ai #llm #openai
Image
This GitHub project caught my eye: bitnet.cpp enables fast, energy-saving inference for 1-bit LLMs (up to 100B params!) on CPUs & GPUs, with tools for benchmarking & troubleshooting. See link below. #AI #LLM #opensource
https://github.com/microsoft/BitNet
Microsoft lance Phi-4 15B : un modèle qui décide lui-même s'il doit "réfléchir" ou répondre du tac au tac. Efficience géniale ou cauchemar de prédictibilité pour les architectes IT ? 🤖⚙️
#AI #Microsoft #Phi4 #Innovation #LLM #CloudComputing
✉️ 𝗦𝗲 𝘃𝘂𝗼𝗶 𝗿𝗶𝗺𝗮𝗻𝗲𝗿𝗲 𝗮𝗴𝗴𝗶𝗼𝗿𝗻𝗮𝘁𝗼/𝗮 𝘀𝘂 𝗾𝘂𝗲𝘀𝘁𝗲 𝘁𝗲𝗺𝗮𝘁𝗶𝗰𝗵𝗲, 𝗶𝘀𝗰𝗿𝗶𝘃𝗶𝘁𝗶 𝗮𝗹𝗹𝗮 𝗺𝗶𝗮 𝗻𝗲𝘄𝘀𝗹𝗲𝘁𝘁𝗲𝗿: bit.ly/newsletter-a...
#AI #GenAI #GenerativeAI #IntelligenzaArtificiale #LLM
🧠 I subagents in #Codex introducono un modo più strutturato di lavorare con l’#AI.
👉 Dettagli e test: www.linkedin.com/posts/alessi...
#AI #GenAI #GenerativeAI #IntelligenzaArtificiale #LLM
More context ≠ better AI coding.
In enterprise Java, a bloated AGENTS.md slows agents down and increases drift.
After using Copilot, Claude, and IBM Bob/Bobshell, I changed how I design repository context.
Full article:
buff.ly/6kpHrsx
#Java #AICoding #LLM #EnterpriseDev
Why your Production Retreival-Augmented-Generation (RAG) is failing and how to fix it? Why Most Retreival-Augmented-Generation (RAG) Systems Fail in Production — and How Developers Can Design Rig...
#design-systems #retrieval-augmented-gen #llm #ai #generative-ai-use-cases
Origin | Interest | […]
