Clinical decision-making is one of the most impactful parts of a physician's responsibilities
and stands to benefit greatly from artificial intelligence solutions and large language models …

Reinforcement learning from human feedback (RLHF) is an effective technique for aligning
large language models (LLMs) to human preferences, but gathering high-quality human …

Generalist foundation models such as GPT-4 have displayed surprising capabilities in a
wide variety of domains and tasks. Yet, there is a prevalent assumption that they cannot …

As research in large language models (LLMs) continues to accelerate, LLM-based
evaluation has emerged as a scalable and cost-effective alternative to human evaluations …

One widely cited barrier to the adoption of LLMs as proxies for humans in subjective tasks is
their sensitivity to prompt wording—but interestingly, humans also display sensitivities to …

Preference learning algorithms (eg, RLHF and DPO) are frequently used to steer LLMs to
produce generations that are more preferred by humans, but our understanding of their …

Reinforcement learning from human feedback (RLHF) has proven effective in aligning large
language models (LLMs) with human preferences, but gathering high-quality preference …

Abstract Generative Artificial Intelligence (GenAI) systems are being increasingly deployed
across all parts of industry and research settings. Developers and end users interact with …

This paper introduces v0. 5 of the AI Safety Benchmark, which has been created by the
MLCommons AI Safety Working Group. The AI Safety Benchmark has been designed to …

Learning with limited labelled data, such as prompting, in-context learning, fine-tuning, meta-
learning, or few-shot learning, aims to effectively train a model using only a small amount of …