Apple’s ReALM ‘sees’ on-screen visuals better than GPT-4
Apple engineers developed an AI system that resolves complex references to on-screen entities and user conversations. The lightweight model could…
Apple reveals MM1, its first family of multimodal LLMs
Apple is yet to officially release an AI model, but a new research paper gives an insight into the company’s…
AI smartphone app can detect depression from facial cues
Researchers from Dartmouth College developed MoodCapture, a smartphone app that uses AI to analyze facial features for early detection of…
Meta releases V-JEPA, a predictive vision model
Meta has released V-JEPA, a predictive vision model that is the next step toward Meta Chief AI Scientist Yann LeCun’s…
V* – Multimodal LLM guided visual search that beats GPT-4V
Researchers from UC San Diego and New York University developed V*, an LLM-guided search algorithm that is a lot better…
Meta releases Ego-Exo4D, a multimodal perception dataset
Training AI models like GPT-4 has relied mostly on datasets consisting of text and images. Meta’s Ego-Exo4D multimodal perception dataset…
Researchers use synthetic data to train AI image classifier
Training an AI image classifier or computer vision system requires access to a large dataset of images. MIT researchers have…
Woodpecker could solve multimodal LLM hallucinations
Multimodal Large Language Models (MLLM) like GPT-4V are really good at analyzing and describing images but sometimes they hallucinate and…
GPT-4’s multimodal capability makes it vulnerable to attack
The ability that GPT-4 has to process images is really impressive, but the new capability opens the model up to…
Machine learning reveals the contents of ancient scrolls and stone tablets
Luke Farritor, a 21-year-old computer science student from the University of Nebraska–Lincoln, has revealed the text inside a carbonized scroll…