A Simple Contrastive Learning Objective for Alleviating Neural Text Degeneration

Shaojie Jiang, Ruqing Zhang, Svitlana Vakulenko, Maarten de Rijke

Keywords: language model, contrastive learning, repetition, degeneration

TL;DR: To tackle the repetitive degeneration problem of neural autoregressive language models, we propose a token-level contrastive learning objective that penalizes incorrectly repeating tokens.

Abstract: The cross-entropy objective has proved to be an all-purpose training objective for autoregressive language models (LMs). However, without distinguishing problematic tokens, LMs trained using cross-entropy exhibit text degeneration problems. To address this, unlikelihood training has been proposed to reduce the probability of unlikely tokens predicted by LMs. But unlikelihood does not explicitly consider the relationship between the label tokens and unlikely token candidates, thus showing marginal improvements in degeneration. We propose a new contrastive token learning objective that inherits the advantages of cross-entropy and unlikelihood training and avoids their limitations. The key idea is to teach a LM to generate high probabilities for label tokens and low probabilities for negative candidates. Comprehensive experiments on language modeling and open-domain dialogue generation tasks show that the proposed contrastive token objective yields much less repetitive texts, with a higher generation quality than baseline approaches, achieving the new state-of-the-art performance on text degeneration.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics

Submission Guidelines: Yes

Please Choose The Closest Area That Your Submission Falls Into: Generative models