What Do You Get When You Cross Beam Search with Nucleus Sampling?

Uri Shaham, Omer Levy

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

We combine beam search with the probabilistic pruning technique of nucleus sampling to create two deterministic nucleus search algorithms for natural language generation. The first algorithm, p-exact search, locally prunes the next-token distribution and performs an exact search over the remaining space. The second algorithm, dynamic beam search, shrinks and expands the beam size according to the entropy of the candidate’s probability distribution. Despite the probabilistic intuition behind nucleus search, experiments on machine translation and summarization benchmarks show that both algorithms reach the same performance levels as standard beam search.

Original languageEnglish
Title of host publicationInsights 2022 - 3rd Workshop on Insights from Negative Results in NLP, Proceedings of the Workshop
EditorsShabnam Tafreshi, Joao Sedoc, Anna Rogers, Aleksandr Drozd, Anna Rumshisky, Arjun Reddy Akula
PublisherAssociation for Computational Linguistics (ACL)
Pages38-45
Number of pages8
ISBN (Electronic)9781955917407
StatePublished - 2022
Event3rd Workshop on Insights from Negative Results in NLP, Insights 2022 - Dublin, Ireland
Duration: 26 May 2022 → …

Publication series

NameInsights 2022 - 3rd Workshop on Insights from Negative Results in NLP, Proceedings of the Workshop

Conference

Conference3rd Workshop on Insights from Negative Results in NLP, Insights 2022
Country/TerritoryIreland
CityDublin
Period26/05/22 → …

Funding

FundersFunder number
Blavatnik Fund
Intel Corporation
Tel Aviv University

    Fingerprint

    Dive into the research topics of 'What Do You Get When You Cross Beam Search with Nucleus Sampling?'. Together they form a unique fingerprint.

    Cite this