New Release
Fine-Tune Wav2Vec2 for English ASR in Hugging Face with 🤗 Transformers
Hugging Face releases a tutorial to fine-tune Wav2Vec2 for improved English ASR performance.
Hugging Face BlogOriginal →
News
2,468 stories
New Release
Hugging Face Reads, Feb. 2021 - Long-range Transformers
Hugging Face's February 2021 update dives into long-range Transformers, enhancing NLP capabilities for lengthy text processing.
Hugging Face BlogOriginal →
New Release
Multimodal neurons in artificial neural networks
New research unveils multimodal neurons in CLIP, enhancing its ability to recognize diverse concepts across various representations.
OpenAI NewsOriginal →
New Release
Simple considerations for simple people building fancy neural networks
Hugging Face shares practical tips to simplify neural network development for beginners.
Hugging Face BlogOriginal →
New Release
Retrieval Augmented Generation with Huggingface Transformers and Ray
Hugging Face and Ray team up to enhance AI capabilities with Retrieval Augmented Generation.
Hugging Face BlogOriginal →
New Release
Hugging Face on PyTorch / XLA TPUs
Hugging Face boosts PyTorch support for XLA TPUs, enhancing performance for large-scale AI models.
Hugging Face BlogOriginal →
Developing
Understanding the capabilities, limitations, and societal impact of large language models
Large language models are reshaping society, but their limitations and biases must be understood for responsible use.
OpenAI NewsOriginal →
New Release
Faster TensorFlow models in Hugging Face Transformers
Hugging Face introduces faster TensorFlow models, enhancing performance and user options in AI applications.
Hugging Face BlogOriginal →
New Release
Scaling Kubernetes to 7,500 nodes
Kubernetes now supports clusters with 7,500 nodes, enhancing scalability for AI applications.
OpenAI NewsOriginal →
New Release
Fit More and Train Faster With ZeRO via DeepSpeed and FairScale
DeepSpeed and FairScale leverage ZeRO technology to enhance model training efficiency, enabling faster and more memory-efficient processes.
Hugging Face BlogOriginal →
New Release
How we sped up transformer inference 100x for 🤗 API customers
Hugging Face announces a groundbreaking 100x speed boost for transformer inference on its API, enhancing user experience.
Hugging Face BlogOriginal →
New Release
CLIP: Connecting text and images
OpenAI's CLIP transforms visual classification by leveraging natural language, enabling efficient image recognition without extensive datasets.
OpenAI NewsOriginal →
New Release
DALL·E: Creating images from text
OpenAI's DALL·E transforms text descriptions into vivid images, showcasing the capabilities of advanced neural networks.
OpenAI NewsOriginal →
New Release
Organizational update from OpenAI
OpenAI announces major organizational changes to enhance efficiency and meet the demands of AI development.
OpenAI NewsOriginal →
New Release
Leveraging Pre-trained Language Model Checkpoints for Encoder-Decoder Models
New research reveals how pre-trained language models can boost encoder-decoder architectures significantly.
Hugging Face BlogOriginal →
New Release
Porting fairseq wmt19 translation system to transformers
Fairseq's WMT19 translation system now operates on the Transformers framework, enhancing performance and compatibility.
Hugging Face BlogOriginal →
New Release
Hyperparameter Search with Transformers and Ray Tune
Hugging Face and Ray Tune join forces to simplify hyperparameter tuning for transformer models.
Hugging Face BlogOriginal →
New Release
Transformer-based Encoder-Decoder Models
Recent advancements in transformer-based encoder-decoder models promise to elevate natural language processing capabilities.
Hugging Face BlogOriginal →
New Release
OpenAI licenses GPT-3 technology to Microsoft
OpenAI has officially licensed its GPT-3 technology to Microsoft, enhancing AI capabilities across Microsoft's product suite.
OpenAI NewsOriginal →
New Release
Block Sparse Matrices for Smaller and Faster Language Models
New block sparse matrix technique slashes language model size and boosts inference speed dramatically.
Hugging Face BlogOriginal →
New Release
Generative language modeling for automated theorem proving
OpenAI's new generative language model boosts automated theorem proving with impressive accuracy and efficiency.
OpenAI NewsOriginal →
New Release
Learning to summarize with human feedback
OpenAI's new models leverage human feedback to significantly enhance summarization capabilities through advanced reinforcement learning techniques.
OpenAI NewsOriginal →
New Release
OpenAI Scholars 2020: Final projects
OpenAI Scholars 2020 showcased innovative AI research projects during a virtual Demo Day, demonstrating practical advancements in the field.
OpenAI NewsOriginal →