Papers
arxiv:2403.13799

Reverse Training to Nurse the Reversal Curse

Published on Mar 20
· Submitted by akhaliq on Mar 21

Abstract

Large language models (LLMs) have a surprising failure: when trained on "A has a feature B", they do not generalize to "B is a feature of A", which is termed the Reversal Curse. Even when training with trillions of tokens this issue still appears due to Zipf's law - hence even if we train on the entire internet. This work proposes an alternative training scheme, called reverse training, whereby all words are used twice, doubling the amount of available tokens. The LLM is trained in both forward and reverse directions by reversing the training strings while preserving (i.e., not reversing) chosen substrings, such as entities. We show that data-matched reverse-trained models provide superior performance to standard models on standard tasks, and compute-matched reverse-trained models provide far superior performance on reversal tasks, helping resolve the reversal curse issue.

Community

This is an automated message from the Librarian Bot. I found the following papers similar to this paper.

The following papers were recommended by the Semantic Scholar API

Please give a thumbs up to this comment if you found it helpful!

If you want recommendations for any Paper on Hugging Face checkout this Space

You can directly ask Librarian Bot for paper recommendations by tagging it in a comment: @librarian-bot recommend

Sign up or log in to comment

Models citing this paper 0

No model linking this paper

Cite arxiv.org/abs/2403.13799 in a model README.md to link it from this page.

Datasets citing this paper 0

No dataset linking this paper

Cite arxiv.org/abs/2403.13799 in a dataset README.md to link it from this page.

Spaces citing this paper 0

No Space linking this paper

Cite arxiv.org/abs/2403.13799 in a Space README.md to link it from this page.

Collections including this paper 4