🚨 Breaking Research Discovery! 🚨 Large Vision Language Models (#VLMs) amaze with coherence but hide risks. 🤯🗡️ 🔍 Meet "Shadowcast"🥷: A stealthy, mind-bending AI data poisoning method. 🕵️‍♂️💻 Project page 🔗: vlm-poison.github.io #LLMs #DataSecurity A 🧵 👇

BLIP3 is here!!!

Excited to share that our newest multi modal foundation model xGen-mm is out and it’s open source! It’s small (<5B models) and shining in both pre-trained and fine-tuned benchmarks. Check it out 👉Hugging Face: tinyurl.com/ybdmf5zs Salesforce AI Research #SalesforceAI #AI #ML

MINT-1T: Scaling Open-Source Multimodal Data by 10x: A Multimodal Dataset with One Trillion Tokens abs: arxiv.org/abs/2406.11271 A new interleaved multimodal pretraining dataset, consists of one trillion text tokens and three billion images, a 10x scale-up from existing

MINT-1T is now available on 🤗 huggingface.co/collections/ml…. A large-scale (1T tokens), open-source, interleaved image-text dataset with diverse data sources (HTML, PDFs, and ArXiv papers).

We are excited to release🍃MINT-1T, the first one trillion token multimodal interleaved dataset with 3.4 billion images, built in collaboration with Salesforce AI Research! Dataset: github.com/mlfoundations/… Paper: arxiv.org/abs/2406.11271 Blog: blog.salesforceairesearch.com/mint-1t/ 🧵

Join us this evening (8 PM PST) for a live discussion on our recent paper and model release 🤗

Excited to share my intern project at Salesforce Research! Huge thanks to everyone on the team!!

We just open sourced TACO 🌮 ! arxiv: arxiv.org/abs/2412.05479 github: github.com/SalesforceAIRe… See this thread to learn more! ⬇️🧵