In-Context Probing: Toward Building Robust Classifiers via Probing Large Language Models
Large language models are able to learn new tasks in context, where they are provided with instructions and a few annotated examples. However, the effectiveness of in-context learning is dependent on the provided context, and the performance on a downstream task can vary considerably, depending on t...
Saved in:
Main Authors | , |
---|---|
Format | Journal Article |
Language | English |
Published |
23.05.2023
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Summary: | Large language models are able to learn new tasks in context, where they are
provided with instructions and a few annotated examples. However, the
effectiveness of in-context learning is dependent on the provided context, and
the performance on a downstream task can vary considerably, depending on the
instruction. Importantly, such dependency on the context can surface in
unpredictable ways, e.g., a seemingly more informative instruction might lead
to a worse performance. In this paper, we propose an alternative approach,
which we term In-Context Probing (ICP). Similar to in-context learning, we
contextualize the representation of the input with an instruction, but instead
of decoding the output prediction, we probe the contextualized representation
to predict the label. Through a series of experiments on a diverse set of
classification tasks, we show that in-context probing is significantly more
robust to changes in instructions. We further show that ICP performs
competitive or superior to finetuning and can be particularly helpful to build
classifiers on top of smaller models, with less than a hundred training
examples. |
---|---|
DOI: | 10.48550/arxiv.2305.14171 |