Contrastive Decoding: Open-ended Text Generation as Optimization
Xiang
Lisa
Li
author
Ari
Holtzman
author
Daniel
Fried
author
Percy
Liang
author
Jason
Eisner
author
Tatsunori
Hashimoto
author
Luke
Zettlemoyer
author
Mike
Lewis
author
2023-07
text
Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
Anna
Rogers
editor
Jordan
Boyd-Graber
editor
Naoaki
Okazaki
editor
Association for Computational Linguistics
Toronto, Canada
conference publication
Given a language model (LM), maximum probability is a poor decoding objective for open-ended generation, because it produces short and repetitive text. On the other hand, sampling can often produce incoherent text that drifts from the original topics. We propose contrastive decoding (CD), a reliable decoding approach that optimizes a contrastive objective subject to a plausibility constraint. The contrastive objective returns the difference between the likelihood under a large LM (called the expert, e.g. OPT-13B) and a small LM (called the amateur, e.g. OPT-125M), and the constraint ensures that the outputs are plausible. CD is inspired by the fact that the failures of larger LMs (e.g., repetition, inco- herence) are even more prevalent in smaller LMs, and that this difference signals which texts should be preferred. CD requires zero additional training, and produces higher quality text than decoding from the larger LM alone. It also works across model scales (OPT-13B and GPT2-1.5B) and significantly outperforms four strong decoding algorithms (e.g., nucleus, top-k) in automatic and human evaluations across wikipedia, news and story domains.
li-etal-2023-contrastive
10.18653/v1/2023.acl-long.687
https://aclanthology.org/2023.acl-long.687
2023-07
12286
12312