LLMs are powerful for reference resolution, including non-conversational entities. This paper shows how LLMs can effectively resolve various references, outperforming existing systems and even GPT-4.

https://arxiv.org/abs//2403.20329

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

TikTok: https://www.tiktok.com/@arxiv_papers

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

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