- AI Research Insights
- Posts
- AI News: HIPAA-compliant LLMs, SenseTime's OmniObject3D, Eye Contact Feature by NVIDIA, VectorMapNet
AI News: HIPAA-compliant LLMs, SenseTime's OmniObject3D, Eye Contact Feature by NVIDIA, VectorMapNet
![](https://media.beehiiv.com/cdn-cgi/image/fit=scale-down,format=auto,onerror=redirect,quality=80/uploads/asset/file/d3225b1b-ba41-4905-a3f0-50779f336b88/Newsletter_2__1_.jpg)
Hi there, today we will share some cool AI research updates such as HIPAA-compliant LLMs, SenseTime's OmniObject3D, Eye Contact Feature by NVIDIA, VectorMapNet, StyleGAN-T, and many more. So, let's start...
SenseTime: Releases a large-vocabulary 3D dataset for realistic perception, reconstruction & generation. Paper/Project, Code & Data (coming). OmniObject3D is a large vocabulary 3D object dataset with massive high-quality real-scanned 3D objects to facilitate the development of 3D perception, reconstruction, and generation in the real world.
HIPAA-compliant LLMs: Microsoft Azure- OpenAI Service is now HIPAA ready (+ includes support for zero-shot or specialized model creation via fine tuning on your own data)!
Introducing Eye Contact by NVIDIA: NVIDIA just released a new Eye Contact feature that uses AI to make you look into the camera. The new Eye Contact effect moves the eyes of the speaker to simulate eye contact with the camera — achieved by estimating and aligning gaze.
Transformer Catalog ( 2023 Edition): Really neat timeline of Transformer models. A great reference to track and find the latest notable Transformer models.
![](https://media.beehiiv.com/cdn-cgi/image/fit=scale-down,format=auto,onerror=redirect,quality=80/uploads/asset/file/cf4f1983-dd76-4086-a44b-0a1043115630/FnAo3Y2XwAEDqZ9__1_.jpeg)
Tsinghua University/MIT: Researchers introduce an end-to-end vectorized HD map learning pipeline, termed VectorMapNet. VectorMapNet takes onboard sensor observations and predicts a sparse set of polylines in the bird’s-eye view. This pipeline can explicitly model the spatial relation between map elements and generate vectorized maps that are friendly to downstream autonomous driving tasks.
StyleGAN-T: Nvidia researchers propose StyleGAN-T. The proposed model addresses the specific requirements of large-scale text-to-image synthesis, such as large capacity, stable training on diverse datasets, strong text alignment, and controllable fidelity vs. text alignment tradeoff. StyleGAN-T significantly improves most of the previous GANs and outperforms distilled diffusion models - the previous state-of-the-art in fast text-to-image synthesis - in terms of sample quality and speed.
Meet GLIGEN: An AI Approach that Extends the Functionality of Existing Pre-Trained Text-to-Image Diffusion Models by Enabling Conditioning on Grounding Inputs
Epoch AI Report: Literature review of Transformative Artificial Intelligence timelines. The research team summarized and compared several models and forecasts predicting when transformative AI will be developed.