SyncAI.news, a Varaisys broadcasting
The Safety Operator: Modulating the Expression of Safety Instructions via Spectral Optimization
BD

Benoit Dherin, Michael Munn, Xavier Gonzalvo, Adrian Goldwaser, Blaz Bratanic, Ananth Balashankar, Andrey Vlasov, Pinzhi Huang, Cecile Loge, Nicole Mitchell, Andre Fernandes, Trilok Acharya, Wendy Kan, Ziyue Wang, Hanna Mazzawi, Felipe Tiengo Ferreira, Mor Geva

· 1 min read

ResearcharXiv cs.AI

The Safety Operator: Modulating the Expression of Safety Instructions via Spectral Optimization

arXiv:2609.36434v1 Announce Type: new Abstract: Context tokens in a transformer-based language model can be absorbed into the model's weights as a multiplicative operator. We study this operator in the setting of safety instructions and show that influencing its dominant eigenvalue modulates how strongly the instruction shapes generation. We derive a Contrastive Safety Loss with a suppression weight that controls the tradeoff between emphasizing the safety instruction on harmful queries while suppressing it on harmless queries. Varying the suppression weight maps a relationship between the attack success and the over-refusal rates, supporting the hypothesis that the operator's eigenvalue acts as a continuous dial for the instruction's influence. Moreover, this relationship holds relatively independently of how the Safety Loss is parameterized, yielding Pareto-improved safety instructions for appropriate values of suppression weight.

Original source

This story was published by arXiv cs.AI and written by Benoit Dherin, Michael Munn, Xavier Gonzalvo, Adrian Goldwaser, Blaz Bratanic, Ananth Balashankar, Andrey Vlasov, Pinzhi Huang, Cecile Loge, Nicole Mitchell, Andre Fernandes, Trilok Acharya, Wendy Kan, Ziyue Wang, Hanna Mazzawi, Felipe Tiengo Ferreira, Mor Geva. SyncAI.news shows a preview; the complete article is on the publisher's site.

Read the full story on arxiv.org

Similar News