Advancing red teaming with people and AI
OpenAI is exploring ways to improve 'red teaming', a method used to test the security of AI systems by simulating attacks. They aim to combine human e...
OpenAI is exploring ways to improve 'red teaming', a method used to test the security of AI systems by simulating attacks. They aim to combine human e...
Researchers have used OpenAI's GPT-4 to fine-tune a computer vision model, improving its ability to create accurate and detailed maps. The technique i...
The Hugging Face team has organized a multilingual debate competition for large language models (LLMs). Five teams participated, each with their own m...
Hugging Face has developed a new method for storing and loading large-scale models, called 'chunking'. This approach breaks down massive files into sm...
Researchers have developed a new technique called self-speculative decoding, which speeds up text generation in language models. This approach allows ...
Hugging Face has introduced an open leaderboard for Japanese language models. The leaderboard allows researchers and developers to compare the perform...
Rox is partnering with OpenAI to use their large language model (LLM) technology. The goal is for sellers using Rox's platform to become top performer...
A new benchmarking system called Judge Arena has been proposed to evaluate the performance of large language models (LLMs) in evaluating other LLMs. T...
OpenAI has opened its first office in continental Europe, located in France. The company's expansion into the region marks a significant step towards ...
The EstΓ©e Lauder Companies is using ChatGPT to gain insights into consumer behavior and preferences in the beauty industry. By analyzing data generate...
Hugging Face, a popular platform for natural language processing and deep learning, is encouraging researchers to share their machine learning (ML) da...
Hugging Face, a popular AI model hub, has integrated its platform with the PyCharm IDE. This integration allows developers to access and use Hugging F...