AI
Controllable Neural Text Generation
Lilian Weng's blog post discusses controllable neural text generation, a technique for generating text that meets specific requirements. The post cove...
AI
AI
Organizational update from OpenAI
OpenAI has announced an organizational update, marking one year since the company's rapid growth and changes. The update highlights key leadership pos...
AI
Leveraging Pre-trained Language Model Checkpoints for Encoder-Decoder Models
Researchers have found a way to improve the performance of encoder-decoder models by leveraging pre-trained language model checkpoints. This approach,...
AI
Porting fairseq wmt19 translation system to transformers
The Hugging Face blog post describes how to port the fairseq WMT'19 translation system to use the Transformers library. The original system was a stat...
AI
Hyperparameter Search with Transformers and Ray Tune
Researchers have developed a method to use transformers, a type of neural network architecture, in conjunction with Ray Tune, an open-source library f...
AI
How to Build an Open-Domain Question Answering System?
Building an open-domain question answering system involves creating a model that can respond to any factual question. This has potential applications ...
AI
Transformer-based Encoder-Decoder Models
Transformer-based encoder-decoder models are a type of neural network architecture that uses self-attention mechanisms to process input sequences. The...
AI
OpenAI licenses GPT-3 technology to Microsoft
OpenAI, a leading AI research organization, has licensed its GPT-3 technology to Microsoft. The agreement allows Microsoft to use the language model i...
AI
Block Sparse Matrices for Smaller and Faster Language Models
Researchers have developed a new technique to optimize language models, called block sparse matrices. This method reduces the number of computations r...
AI
Generative language modeling for automated theorem proving
Researchers at OpenAI have applied generative language models to automate theorem proving, a task typically performed by humans. They used the model t...
AI
Learning to summarize with human feedback
Researchers have used a technique called reinforcement learning from human feedback to improve the ability of language models to summarize text. This ...
AI
Neural Architecture Search
Researchers are exploring ways to automatically design neural networks, rather than relying on human experts. This approach is called Neural Architect...