TL;DR
This paper introduces a model-free probing method using prompting to analyze linguistic information in language models, demonstrating comparable or superior extraction with less learning, and explores the architecture's storage of linguistic info.
Contribution
It proposes a novel prompting-based probing approach that avoids model-dependent diagnostics and combines it with attention pruning to analyze linguistic information storage.
Findings
Prompting-based probing is as effective or better than diagnostic probes.
Pruning attention heads affects the model's linguistic property encoding.
Removing heads critical for a property impacts language modeling performance.
Abstract
Probing is a popular method to discern what linguistic information is contained in the representations of pre-trained language models. However, the mechanism of selecting the probe model has recently been subject to intense debate, as it is not clear if the probes are merely extracting information or modeling the linguistic property themselves. To address this challenge, this paper introduces a novel model-free approach to probing, by formulating probing as a prompting task. We conduct experiments on five probing tasks and show that our approach is comparable or better at extracting information than diagnostic probes while learning much less on its own. We further combine the probing via prompting approach with attention head pruning to analyze where the model stores the linguistic information in its architecture. We then examine the usefulness of a specific linguistic property for…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Code & Models
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
MethodsPruning
