When trying to gain better visibility into a machine learning model in o...
Work on scaling laws has found that large language models (LMs) show
pre...
Large Language Models (LLMs) can achieve strong performance on many task...
Pretrained language models often generate outputs that are not in line w...
The potential for pre-trained large language models (LLMs) to use natura...
Language models (LMs) are pretrained to imitate internet text, including...
We test the hypothesis that language models trained with reinforcement
l...
As AI systems become more capable, we would like to enlist their help to...
Developing safe and useful general-purpose AI systems will require us to...
We describe our early efforts to red team language models in order to
si...
Prior work on language models (LMs) shows that training on a large numbe...
We study whether language models can evaluate the validity of their own
...
Reinforcement learning (RL) is frequently employed in fine-tuning large
...
Pretrained language models often do not perform tasks in ways that are i...
Current QA systems can generate reasonable-sounding yet false answers wi...
Language Models (LMs) often cannot be deployed because of their potentia...
Pretrained language models (LMs) perform well on many tasks even when
le...
It is often challenging for a system to solve a new complex problem from...
We introduce a method to determine if a certain capability helps to achi...
Large pre-trained language models have been shown to store factual knowl...
We aim to improve question answering (QA) by decomposing hard questions ...
We propose a system that finds the strongest supporting evidence for a g...
We introduce the first large-scale corpus for long-form question answeri...
Recent breakthroughs in computer vision and natural language processing ...
We introduce HoME: a Household Multimodal Environment for artificial age...
We introduce a general-purpose conditioning method for neural networks c...
Achieving artificial visual reasoning - the ability to answer image-rela...
Semi-supervised learning algorithms reduce the high cost of acquiring la...