Neural Text Generation with Unlikelihood Training

08/12/2019
by   Sean Welleck, et al.
2

Neural text generation is a key tool in natural language applications, but it is well known there are major problems at its core. In particular, standard likelihood training and decoding leads to dull and repetitive responses. While some post-hoc fixes have been proposed, in particular top-k and nucleus sampling, they do not address the fact that the token-level probabilities predicted by the model itself are poor. In this paper we show that the likelihood objective itself is at fault, resulting in a model that assigns too much probability to sequences that contain repeats and frequent words unlike the human training distribution. We propose a new objective, unlikelihood training, which forces unlikely generations to be assigned lower probability by the model. We show that both token and sequence level unlikelihood training give less repetitive, less dull text while maintaining perplexity, giving far superior generations using standard greedy or beam search. Our approach provides a strong alternative to traditional training.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
02/13/2022

A Contrastive Framework for Neural Text Generation

Text generation is of great importance to many natural language processi...
research
05/22/2020

Investigating Label Bias in Beam Search for Open-ended Text Generation

Beam search is an effective and widely used decoding algorithm in many s...
research
06/08/2023

SequenceMatch: Imitation Learning for Autoregressive Sequence Modelling with Backtracking

In many domains, autoregressive models can achieve low log-likelihood on...
research
03/13/2021

Improving Diversity of Neural Text Generation via Inverse Probability Weighting

The neural network based text generation suffers from the text degenerat...
research
02/26/2023

Tailoring Language Generation Models under Total Variation Distance

The standard paradigm of neural language generation adopts maximum likel...
research
09/30/2022

Calibrating Sequence likelihood Improves Conditional Language Generation

Conditional language models are predominantly trained with maximum likel...
research
01/20/2021

Generating (Formulaic) Text by Splicing Together Nearest Neighbors

We propose to tackle conditional text generation tasks, especially those...

Please sign up or login with your details

Forgot password? Click here to reset