The Inductive Bias of In-Context Learning: Rethinking Pretraining Example Design

by   Yoav Levine, et al.

Pretraining Neural Language Models (NLMs) over a large corpus involves chunking the text into training examples, which are contiguous text segments of sizes processable by the neural architecture. We highlight a bias introduced by this common practice: we prove that the pretrained NLM can model much stronger dependencies between text segments that appeared in the same training example, than it can between text segments that appeared in different training examples. This intuitive result has a twofold role. First, it formalizes the motivation behind a broad line of recent successful NLM training heuristics, proposed for the pretraining and fine-tuning stages, which do not necessarily appear related at first glance. Second, our result clearly indicates further improvements to be made in NLM pretraining for the benefit of Natural Language Understanding tasks. As an example, we propose "kNN-Pretraining": we show that including semantically related non-neighboring sentences in the same pretraining example yields improved sentence representations and open domain question answering abilities. This theoretically motivated degree of freedom for "pretraining example design" indicates new training schemes for self-improving representations.



There are no comments yet.


page 1

page 2

page 3

page 4


Back-Translated Task Adaptive Pretraining: Improving Accuracy and Robustness on Text Classification

Language models (LMs) pretrained on a large text corpus and fine-tuned o...

The Effectiveness of Intermediate-Task Training for Code-Switched Natural Language Understanding

While recent benchmarks have spurred a lot of new work on improving the ...

Collecting Entailment Data for Pretraining: New Protocols and Negative Results

Textual entailment (or NLI) data has proven useful as pretraining data f...

On a Benefit of Mask Language Modeling: Robustness to Simplicity Bias

Despite the success of pretrained masked language models (MLM), why MLM ...

Few-Shot Question Answering by Pretraining Span Selection

In a number of question answering (QA) benchmarks, pretrained models hav...

Does Pretraining for Summarization Require Knowledge Transfer?

Pretraining techniques leveraging enormous datasets have driven recent a...

Improving Code Autocompletion with Transfer Learning

Software language models have achieved promising results predicting code...
This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.