Trolling GPT4 It's remarkable still how gullible ChatGPT can be, let's see what we can extract from it... 🧵 A thread

𝐎𝐏𝐓𝐀 𝐅𝐎𝐑𝐔𝐌 𝟐𝟎𝟐𝟑 Earlier this year Guillaume Hacques explained how an analyst can use tracking data to measure the collective moments of an attacking team, when facing a set defence, based on how effective the movement is at causing shape disruption. vimeo.com/817400097

Après Mac Kinsey, Mac Vacant. 💨 Ou quand la Macronie utilise des cabinets de Conseil pour apprendre à faire des courriers… La vérité sur les cabinets de conseil : ne ratez pas la fin de la video !

REAL-TIME object detection WITHOUT TRAINING YOLO-World is a new SOTA open-vocabulary object detector that outperforms previous models in terms of both accuracy and speed. 35.4 AP with 52.0 FPS on V100. ↓ read more

Watch Gemini 1.5 Pro collect all my tweets from a video where I scroll through my feed, and neatly organize them into a JSON format. It's even smart enough to discern between tweets, retweets, and quotes. Best scraper tool on the planet! ⚡️

This is mind blowing. This AI can make single image sing, talk, and rap from any audio file expressively! 🤯 Introducing EMO: Emote Portrait Alive by Alibaba. 10 wild examples: 🧵👇 1. AI Lady from Sora singing Dua Lipa

In the era of long-context LLMs, I needed a way to quickly convert a GitHub repo into a text file to feed as a prompt. Introducing RepoToTextForLLMs 📖 Not only does it return the full repo in one file, but it also appends a super prompt for analysis and understanding 👇

Today we're excited to introduce Devin, the first AI software engineer. Devin is the new state-of-the-art on the SWE-Bench coding benchmark, has successfully passed practical engineering interviews from leading AI companies, and has even completed real jobs on Upwork. Devin is

i never believe recorded demos so I reached out to the @cognition team for early access to try for myself and got it! will be sharing my unfiltered opinions on #devin here. 🧵🧵 1/n

YOLO v9 is out! The state of the art in real-time object detection! It's faster & more accurate than any previous version of YOLO, beating all convolutional or transformer-based models. It comes with two new feature: - Programmable Gradient Information (PGI), which allows YOLO

BILLET. Dans «Touche pas à mon poste», le député macroniste Quentin Bataillon a dénigré Yann Barthès, auditionné par sa commission. Il a perdu toute crédibilité et doit quitter son poste. Par thomas legrand ⤵️ liberation.fr/politique/comm…

PyTorch stepping into the LLM-fine-tuning arena feels huge, looking forward to seeing torchtune evolve in the coming months Tutorials and docs are really nicely built with a blend of education (eg LoRA explainer) as well as how-tos wandb integration the cherry on top 👇

This is not offside. x.com/balltopmb/stat…

🚨 OpenAI has just posted about "Shaping Desired Model Behavior," and I strongly disagree with the proposed rule below, as it's a slippery slope for dangerous misinformation.

Want to paste your entire codebase to Chatgpt, Claude or your local LLM of choice including folder structure? Use the small CLI tools code2prompt! Have been using it for some time and it works very well!

arXiv -> alphaXiv Students at Stanford have built alphaXiv, an open discussion forum for arXiv papers. alphaXiv You can post questions and comments directly on top of any arXiv paper by changing arXiv to alphaXiv in any URL!

Accurate.

introducing Open NotebookLM turn any PDF 📄 into a personalized podcast 🎧 in no time. the best part? it was all built in a single afternoon using open-source AI ✨. 🧵 (1/4)

SmolVLM-500M from Hugging Face is a tiny multimodal model but its just so useful. Here's a webcam demo in real-time powered SmolVLM and llama.cpp, all local on an M3 Macbook.