Papers
arxiv:2401.17181

Transfer Learning for Text Diffusion Models

Published on Jan 30
· Submitted by akhaliq on Jan 31
Authors:
,
,

Abstract

In this report, we explore the potential for text diffusion to replace autoregressive (AR) decoding for the training and deployment of large language models (LLMs). We are particularly interested to see whether pretrained AR models can be transformed into text diffusion models through a lightweight adaptation procedure we call ``AR2Diff''. We begin by establishing a strong baseline setup for training text diffusion models. Comparing across multiple architectures and pretraining objectives, we find that training a decoder-only model with a prefix LM objective is best or near-best across several tasks. Building on this finding, we test various transfer learning setups for text diffusion models. On machine translation, we find that text diffusion underperforms the standard AR approach. However, on code synthesis and extractive QA, we find diffusion models trained from scratch outperform AR models in many cases. We also observe quality gains from AR2Diff -- adapting AR models to use diffusion decoding. These results are promising given that text diffusion is relatively underexplored and can be significantly faster than AR decoding for long text generation.

Community

I get so excited thinking abouting the possibility of diffusion based long-context code generation. It seems like it could be a match made in heaven, with error detection built in to the denoising stages. Now add a control net in to the mix to add new features 🙏 Imagine how fast we could build software when this is a reality. The average programmer could build an entire operating system and a full stack from scratch in a few hours

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

IMG_8302.png

Sign up or log in to comment

Models citing this paper 0

No model linking this paper

Cite arxiv.org/abs/2401.17181 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/2401.17181 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/2401.17181 in a Space README.md to link it from this page.

Collections including this paper 7