Teaching with AI
OpenAI releases a teacher's guide for using ChatGPT in classrooms. It covers suggested prompts, how ChatGPT works and its limitations, the effectiveness of AI detectors, and bias issues.
23 articles
OpenAI releases a teacher's guide for using ChatGPT in classrooms. It covers suggested prompts, how ChatGPT works and its limitations, the effectiveness of AI detectors, and bias issues.
Hugging Face releases an optimized version of AudioLDM 2 for audio generation. Speed improvement is the main focus, though specific performance metrics are not provided in the excerpt.
OpenAI launches ChatGPT Enterprise, a version with enterprise-grade security and privacy. Delivers the most powerful ChatGPT capabilities to date.
Hugging Face deprecates Git password authentication. Users must migrate to personal access tokens or SSH keys before the deadline. This change enhances repository security.
Meta releases Code Llama, a Llama 2 variant specialized in code generation. The model supports multiple programming languages and delivers performance comparable to existing code generation tools for completion and generation tasks.
OpenAI partners with Scale to help enterprises fine-tune its most advanced models. OpenAI customers gain access to Scale's AI expertise to customize models.
Hugging Face introduces AutoGPTQ, a quantization method to reduce LLM size. Integration into the transformers library enables model compression while maintaining performance, easing deployment on resource-constrained hardware.
OpenAI enables fine-tuning of GPT-3.5 Turbo, allowing developers to customize the model with their own data for specific use cases.
Hugging Face introduces SafeCoder, a code security tool. The system analyzes and validates generated code to detect vulnerabilities and bad practices before deployment.
Hugging Face introduces IDEFICS, an open-source reproduction of a state-of-the-art visual language model. The model combines vision and language for multimodal tasks, with code and weights publicly available.
OpenAI acquires Global Illumination. The entire team has joined OpenAI.
OpenAI uses GPT-4 for content policy development and moderation decisions. The approach reduces human moderator workload, accelerates policy refinement feedback loops, and improves labeling consistency.
Hugging Face Hub is now available on AWS Marketplace. Users can access models and datasets through their existing AWS account with integrated billing.
Hugging Face and BentoML demonstrate deploying DeepFloyd IF (image generation model) using BentoML. Practical guide covering containerization, scalability, and integration with Hugging Face ecosystem for production.
Hugging Face optimizes Bark, its text-to-speech model, using the Transformers library. Improvements include reduced latency, memory compression, and better audio quality. Optimized code and models available on the hub.
Hugging Face releases Swift Transformers, a library to run LLMs directly on Apple devices. Compatible with Core ML and optimized for on-device performance, it enables model deployment without network connectivity.
Hugging Face publishes a guide to fine-tune Llama 2 using DPO (Direct Preference Optimization). The method aligns the model to user preferences without explicit reward modeling, reducing computational costs compared to traditional RLHF approaches.
Hugging Face enables streamlined MusicGen deployment through Inference Endpoints. Users can generate music without infrastructure management, with Meta model support and direct API integration.
Hugging Face introduces Huggy Lingo, an ML system to improve language metadata on its Hub. The tool automatically detects languages in models and datasets, enriching catalog information to improve discoverability and interoperability.
Hugging Face explores fully homomorphic encryption (FHE) for large language models, enabling inference on encrypted data without decryption. Experimental approach aimed at protecting user privacy during LLM processing.
OpenAI publishes workshop proceedings on confidence-building measures for artificial intelligence. The document explores verification frameworks, transparency protocols, and governance mechanisms to strengthen trust in AI systems.
Practical step-by-step guide for generating 3D assets. Covers tools, workflows, and best practices for creating 3D models with AI.
Hugging Face open-sources knowledge distillation code and weights for SD-Small and SD-Tiny, two compressed diffusion models. These variants reduce model size and latency while preserving image generation quality.