Tell, Don't Show: Internalized Reasoning influences how LLMs generalize
Primary Area: generative models
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: LLM generalization, AI safety
Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2024/AuthorGuide.
TL;DR: When training an LLM on examples that follow a pattern, but also giving the model declarative information stating the opposite of the pattern at train time, we investigate how the model ends up generalizing.
Abstract: In this paper we investigate to what extent language models' generalization behavior during a domain shift can be influenced by declarative knowledge contained in the training data.
In order to study this we finetune language models to fit some distribution which has a ``natural'' generalization when the distribution shifts.
We then test to what extent declarative statements in the training data - that if fully internalized would greatly affect the domain shift generalization - can indeed alter the model's behavior on unseen examples.
While the effect is subtle, the declarative knowledge provided in the finetuning sets systematically changes the models' predictions in the way one would expect.
Evidence for the strength of this effect growing with model size is mixed.
We further show that the effect can not be explained by simple token matching behavior as it persists even when there is no overlap between the declarative descriptions and the models' test time generations.
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: 4981
Loading