$C$-$\Delta\Theta$: Circuit-Restricted Weight Arithmetic for Selective Refusal
Mirrored from arXiv — NLP / Computation & Language for archival readability. Support the source by reading on the original site.
Computer Science > Computation and Language
Title:$C$-$ΔΘ$: Circuit-Restricted Weight Arithmetic for Selective Refusal
Abstract:Modern deployments require LLMs to enforce safety policies at scale, yet many controls rely on inference-time interventions that add recurring compute cost and serving complexity. Activation steering is widely used, but it requires runtime hooks and scales cost with the number of generations; conditional variants improve selectivity by gating when steering is applied but still retain an inference-time control path. We ask whether selective refusal can be moved entirely offline: can a mechanistic understanding of category-specific refusal be distilled into a circuit-restricted weight update that deploys as a standard checkpoint? We propose C-{\Delta}{\theta} Circuit Restricted Weight Arithmetic}, which (i) localizes refusal-causal computation as a sparse circuit using EAP-IG and (ii) computes a constrained weight update {\Delta}{\theta}C supported only on that circuit (typically <5% of parameters). Applying {\Delta}{\theta}C yields a drop-in edited checkpoint with no inference-time hooks, shifting cost from per request intervention to a one-time offline update. We evaluate category-targeted selectivity and capability retention on refusal and utility benchmarks.
| Subjects: | Computation and Language (cs.CL); Emerging Technologies (cs.ET) |
| Cite as: | arXiv:2602.04521 [cs.CL] |
| (or arXiv:2602.04521v2 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2602.04521
arXiv-issued DOI via DataCite
|
Submission history
From: Pratinav Seth [view email][v1] Wed, 4 Feb 2026 13:10:52 UTC (5,063 KB)
[v2] Wed, 8 Jul 2026 17:47:06 UTC (2,649 KB)
Access Paper:
- View PDF
- HTML (experimental)
- TeX Source
References & Citations
Bibliographic and Citation Tools
Code, Data and Media Associated with this Article
Demos
Recommenders and Search Tools
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
More from arXiv — NLP / Computation & Language
-
Recipes for Steering and Scaling LLMs via Sampling
Aug 28
-
Can a Model Catch Its Own Hallucinations for Free?: Label-Free Doubt Signals Hold Their Own Against a Labelled Dataset for Abstention
Aug 28
-
FIRSTPASS: A Multi-Domain, Multi-Round Peer Review Dataset Grounded in Real Editorial Outcomes
Aug 28
-
Interpretable, Fairly Evaluated Automated L2 Speaking Assessment that Beats the Single-Human Ceiling and Why Pause Encoding Does Not Change LLM Fluency Scores
Aug 28
Discussion (0)
Sign in to join the discussion. Free account, 30 seconds — email code or GitHub.
Sign in →No comments yet. Sign in and be the first to say something.