When Machine Learning technologies are used in contexts that affect citizens, companies as
well as researchers need to be confident that there will not be any unexpected social …

In this work, we survey a breadth of literature that has revealed the limitations of
predominant practices for dataset collection and use in the field of machine learning. We …

We report the development of GPT-4, a large-scale, multimodal model which can accept
image and text inputs and produce text outputs. While less capable than humans in many …

Aligning large language models (LLMs) with human preferences has proven to drastically
improve usability and has driven rapid adoption as demonstrated by ChatGPT. Alignment …

Large language models are now tuned to align with the goals of their creators, namely to be"
helpful and harmless." These models should respond helpfully to user questions, but refuse …

As the capabilities of generative language models continue to advance, the implications of
biases ingrained within these models have garnered increasing attention from researchers …

Large language models (LLMs), exemplified by ChatGPT, have gained considerable
attention for their excellent natural language processing capabilities. Nonetheless, these …

Responsible innovation on large-scale Language Models (LMs) requires foresight into and
in-depth understanding of the risks these models may pose. This paper develops a …

Large language models have been shown to achieve remarkable performance across a
variety of natural language tasks using few-shot learning, which drastically reduces the …

Large language models (LLMs) have gained considerable attention for their excellent
natural language processing capabilities. Nonetheless, these LLMs present many …