Large Language Models (MLLMs) to follow task instructions and improve generalization ability.

How to evaluate Large Language Models (LLMs) in code generation remains an open question.

This intervention significantly improves the performance of LLaMA models on the TruthfulQA benchmark.

We introduce the Cambridge Law Corpus (CLC), a corpus for legal AI research.

Text watermarks for large language models (LLMs) have been commonly used to identify the origins of machine-generated content, which is promising for assessing liability when combating deepfake or harmful content.