Programming Refusal with Conditional Activation Steering

ICLR 2025 Conference Submission1937 Authors

Keywords: Activation Engineering, Safety, Alignment, Steering Vector

TL;DR: Conditional Activation Steering enables context-dependent control over LLM behaviors by conditionally applying activation steering based on input context during inference without weight optimization.

Abstract:

LLMs have shown remarkable capabilities, but precisely controlling their response behavior remains challenging.
Existing activation steering methods alter LLM behavior indiscriminately, limiting their practical applicability in settings where selective responses are essential, such as content moderation or domain-specific assistants.
In this paper, we propose Conditional Activation Steering (CAST), which analyzes LLM activation patterns during inference to selectively apply or withhold activation steering based on the input context.
Our method is based on the observation that different categories of prompts activate distinct patterns in the model's hidden states.
Using CAST, one can systematically control LLM behavior with rules like "if input is about hate speech or adult content, then refuse" or "if input is not about legal advice, then refuse."
This allows for selective modification of responses to specific content while maintaining normal responses to other content, all without requiring weight optimization.
We release an open-source implementation of our framework at <placeholder: open-source GitHub link>.

Primary Area: alignment, fairness, safety, privacy, and societal considerations

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.

Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2025/AuthorGuide.

Reciprocal Reviewing: I understand the reciprocal reviewing requirement as described on https://iclr.cc/Conferences/2025/CallForPapers. If none of the authors are registered as a reviewer, it may result in a desk rejection at the discretion of the program chairs. To request an exception, please complete this form at https://forms.gle/Huojr6VjkFxiQsUp6.

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: 1937