← Search

Nathan Zixia Hu

2 accepted papers

2024

Long-form factuality in large language models

NeurIPS 2024poster

Large language models (LLMs) often generate content that contains factual errors when responding to fact-seeking prompts on open-ended topics. To benchmark a model’s long-form factuality in open domains, we first use GPT-4 to generate LongFact, a prompt set comprising thousands of questions spanning…

2023

Meta-Learning Online Adaptation of Language Models

EMNLP 2023long main

Large language models encode impressively broad world knowledge in their parameters. However, the knowledge in static language models falls out of date, limiting the model’s effective “shelf life.” While online fine-tuning can reduce this degradation, we find that naively fine-tuning on a stream of…

Cited by 0SourcecodeScholar