Gradient Ascent Post-training Enhances Language Model Generalization
Dongkeun Yoon, Joel Jang, Sungdong Kim, Minjoon Seo
Anthology
Underline
Main: Large Language Models Main-poster Paper
Session 7: Large Language Models (Virtual Poster)
Conference Room: Pier 7&8
Conference Time: July 12, 11:00-12:30 (EDT) (America/Toronto)
Global Time: July 12, Session 7 (15:00-16:30 UTC)
Keywords:
robustness, fine-tuning
TLDR:
In this work, we empirically show that updating pretrained LMs (350M, 1.3B, 2.7B) with just a few steps of Gradient Ascent Post-training (GAP) on random, unlabeled text corpora enhances its zero-shot generalization capabilities across diverse NLP tasks. Specifically, we show that GAP can allow LMs t...
You can open the
#paper-P3294
channel in a separate window.
Abstract:
In this work, we empirically show that updating pretrained LMs (350M, 1.3B, 2.7B) with just a few steps of Gradient Ascent Post-training (GAP) on random, unlabeled text corpora enhances its zero-shot generalization capabilities across diverse NLP tasks. Specifically, we show that GAP can allow LMs to become comparable to 2-3x times larger LMs across 12 different NLP tasks. We also show that applying
GAP on out-of-distribution corpora leads to the most reliable performance improvements. Our findings indicate that GAP can be a promising method for improving the generalization capability of LMs without any task-specific fine-tuning.