Widely used alignment techniques, such as reinforcement learning from human feedback
(RLHF), rely on the ability of humans to supervise model behavior-for example, to evaluate …

When large language models (LLMs) exceed human-level capabilities, it becomes
increasingly challenging to provide full-scale and accurate supervision for these models …

The burgeoning capabilities of large language models (LLMs) have underscored the need
for alignment to ensure these models act in accordance with human values and intentions …

Large language models (LLMs) can solve arithmetic word problems with high accuracy, but
little is known about how well they generalize to problems that are more complex than the …

How can we harness the collective capabilities of multiple Large Language Models (LLMs)
to create an even more powerful model? This question forms the foundation of our research …

Understanding and mitigating the potential risks associated with foundation models (FMs)
hinges on developing effective interpretability methods. Sparse Autoencoders (SAEs) have …

Large language models (LLMs) have demonstrated impressive capabilities, but still suffer
from inconsistency issues (eg LLMs can react differently to disturbances like rephrasing or …

While generalization over tasks from easy to hard is crucial to profile language models
(LLMs), the datasets with fine-grained difficulty annotations for each problem across a broad …

Aligning large language models (LLMs) to human preferences is challenging in domains
where preference data is unavailable. We address the problem of learning reward models …

Fine-tuning Large Language Models (LLMs) incurs considerable training costs, driving the
need for data-efficient training with optimised data ordering. Human-inspired strategies offer …