Introducing ChatGPT
OpenAI introduces ChatGPT, a model trained to interact conversationally. The dialogue format enables ChatGPT to answer follow-up questions, admit mistakes, challenge incorrect premises, and reject inappropriate requests.
11 articles
OpenAI introduces ChatGPT, a model trained to interact conversationally. The dialogue format enables ChatGPT to answer follow-up questions, admit mistakes, challenge incorrect premises, and reject inappropriate requests.
VQ-Diffusion is an image generation model combining vector quantization and diffusion. It uses a discrete codebook to represent images, enabling more efficient and controllable generation than standard diffusion approaches.
Hugging Face outlines its inference solutions: Inference API (serverless), Dedicated Endpoints (dedicated instances), Text Generation Inference (TGI, LLM-optimized), and Transformers.js (browser inference). Each solution addresses different use cases based on latency, cost, and scalability requirements.
Hugging Face accelerates document AI capabilities with new optimized models and tools for document processing. The platform improves inference speed and accuracy on tasks like information extraction and document classification.
Hugging Face demonstrates sentiment analysis on encrypted data using homomorphic encryption. The technique enables processing text without decryption, preserving privacy while leveraging standard NLP models.
Hugging Face launches interactive ML demos directly on arXiv, enabling readers to test models and paper results without leaving the platform. Native integration of Hugging Face Spaces into scientific articles.
Hugging Face introduces contrastive search, a decoding method for transformers that generates human-level text. The technique balances likelihood and lexical diversity, improving quality over standard beam search and sampling approaches.
Hugging Face releases a comprehensive guide for training Stable Diffusion with DreamBooth using the Diffusers library. The method enables model personalization on specific images through resource-efficient fine-tuning.
OpenAI makes DALL·E API available in public beta. Developers can now build applications using the image generation model.
Hugging Face releases a guide to fine-tune Whisper, OpenAI's speech recognition model, on multilingual data using the Transformers library. The method covers audio preprocessing, training, and evaluation across multiple languages.
Hugging Face announces integration of Optimum Intel and OpenVINO to accelerate model inference. The solution optimizes performance on Intel processors through quantization and compilation.