Papers by Widyadewi Soedarmadji

1 papers
Virtual Personas for Language Models via an Anthology of Backstories (2024.emnlp-main)

Copied to clipboard

Challenge: Large language models (LLMs) are trained from vast repositories of text authored by millions of distinct authors, reflecting an enormous diversity of human traits.
Approach: They propose a method for conditioning LLMs to particular virtual personas by harnessing open-ended life narratives, which they refer to as backstories, and demonstrate that it improves consistency and reliability of experimental outcomes.
Outcome: The proposed method improves consistency and reliability of experimental outcomes while ensuring better representation of diverse sub-populations.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations