Perturbed examples reveal invariances shared by language models
Primary Area: visualization or interpretation of learned representations
Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.
Keywords: behavioral comparison, linguistic capabilities, shared invariance
Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2024/AuthorGuide.
TL;DR: A framework for direct behavioral comparison between two NLP models by evaluating the degree to which a target model shares behavioral similarity (and invariances) on a linguistic capability defined with respect to a reference model.
Abstract: An explosion of work in language is leading to ever-increasing numbers of available natural language processing models, with little understanding of how new models compare to better-understood models. One major reason for this difficulty is saturating benchmark datasets, which may not reflect well differences in model performance in the wild. In this work, we propose a novel framework for comparing two natural language processing models by revealing their shared invariance to interpretable input perturbations that are designed to target a specific linguistic capability (e.g., Synonym-Invariance, Typo-Invariance). Via experiments on models from within the same and across different architecture families, this framework offers a number of insights about how changes in models (e.g. distillation, increase in size, amount of pre-training) affect multiple well-defined linguistic capabilities. Furthermore, we also demonstrate how our framework can enable evaluation of the invariances shared between models that are available as commercial black-box APIs (e.g., InstructGPT family) and models that are relatively better understood (e.g., GPT-2). Across several experiments, we observe that large language models share many of the invariances encoded by models of various sizes, whereas the invariances encoded by large language models are only shared by other large models. Possessing a wide variety of invariances may be a key reason for the recent successes of large language models, and our framework can shed light on the types of invariances that are retained by or emerge in new models.
Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors' identity.
No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.
Submission Number: 7449
Loading