The paper proposes a decoding strategy called DoLa to reduce hallucinations in large language models (LLMs) without the need for external knowledge or additional fine-tuning. DoLa improves truthfulness in LLMs across multiple tasks.

https://arxiv.org/abs//2309.03883

YouTube: https://www.youtube.com/@ArxivPapers

PODCASTS:

Apple Podcasts: https://podcasts.apple.com/us/podcast/arxiv-papers/id1692476016

Spotify: https://podcasters.spotify.com/pod/show/arxiv-papers