If you are interested in drone detection techniques, check this 38-page paper on RF-tech for spotting, classifying & tracking drones using radar and comms. Link in the comment.

The whitepaper is live! Learn how to win the HTTP desync endgame... and why HTTP/1.1 needs to die: http1mustdie.com

We are releasing NuMarkdown-8B-Thinking, an open-source (MIT License) reasoning OCR model 🧠✨📄 NuMarkdown-8B-Thinking is apparently the first (!) reasoning VLM specialized in converting PDFs/Scans/Spreadsheets into Markdown files (typically used for RAG applications). It

recently found out about 2 scaling books: 1. tpu jax-ml.github.io/scaling-book/ 2. gpu huggingface.co/spaces/nanotro… I imagine that for me despite not having several accel. to train on, it's a good thing to know these dem rules

whisper.cpp is coming to ffmpeg github.com/FFmpeg/FFmpeg/…

Blog post with examples by the author of the patch: medium.com/@vpalmisano/ru…

Google just released a python library for data extraction! LangExtract is a python library that extracts structured information from unstructured text documents with precise source grounding and interactive visualization. 100% Open Source

Introducing Gemma 3 270M 🔥 🤏A tiny model! Just 270 million parameters 🧠 Very strong instruction following 🤖 Fine-tune in just a few minutes, with a large vocabulary to serve as a high-quality foundation developers.googleblog.com/en/introducing…

🚨 Big news! We decided that Hugging Face’s post-training library, TRL, will natively supports training Vision Language Models 🖼️ This builds on our recent VLM support in SFTTrainer — and we’re not stopping until TRL is the #1 VLM training library 🥇 More here 👉

🚀 Amazing community project! vLLM CLI — a command-line tool for serving LLMs with vLLM: ✅ Interactive menu-driven UI & scripting-friendly CLI ✅ Local + HuggingFace Hub model management ✅ Config profiles for perf/memory tuning ✅ Real-time server & GPU monitoring ✅ Error

today we are teaching a machine to exploit object detection models using swarm drones + reinforcement learning after only 5,000 steps on my Mac Air M3, the policy brought down detection from 90% to 32%, allowing 20+ drones to fly undetected

Learn to fine-tune OpenAI gpt-oss with our new step-by-step guide! Learn about: • Local gpt-oss training + inference FAQ & tips • Evaluation, hyperparameters & overfitting • Reasoning effort, Data prep • Run & saving your LLM to llama.cpp GGUF, HF 🔗docs.unsloth.ai/basics/tutoria…

Most LLM deployments today waste compute & money. In this video, Mark Kurtz explains why models outpace hardware, how compression (quantization, pruning, distillation) works, and how optimized models can cut costs by up to 8x. youtube.com/watch?v=q6BotP…

Today we're releasing NVIDIA Nemotron Nano v2 - a 9B hybrid SSM that is 6X faster than similarly sized models, while also being more accurate. Along with this model, we are also releasing most of the data we used to create it, including the pretraining corpus. Links to the

Teknium (e/λ) This is a "runtime" feature. We started with same approach as Qwen3 but noticed that the model starts "thinking" outside of the thinking trace of forced to answer. Training on truncated thinking traces fixed that. Section 3.4 research.nvidia.com/labs/adlr/file…

At long last - Phrack 72 has been released online for your reading pleasure! Check it out: phrack.org

Over the past few months, CrowdStrike blocked a sophisticated malware campaign that targeted over 300 customer environments. The campaign deployed SHAMOS, a variant of Atomic macOS Stealer developed by COOKIE SPIDER Read more here: crowdstrike.com/en-us/blog/fal…

Today's AI Agent architectures (ReAct, Plan-then-Act, etc) produce too many incorrect responses. Our new benchmark confirms this, evaluating 5 popular Agent architectures in multi-hop Question-Answering. We then added real-time trust scoring to each one, which reduced

Update to iOS 18.6.2 to block what is likely a 0-click. Extremely sophisticated = the current standard. Dear 🍎: Specific targeted individuals? Give us rough numbers, sector, segments, anything… My take is that this is a drop an in ocean of zero clicks.

just a minor version bump. booooring