SafeSteer restricts reverse KL penalty to safety tokens selected via activation steering, achieving strong safety on seven benchmarks with minimal degradation on five capability benchmarks using only 100 harmful samples and no general data.
Mitigating the safety alignment tax with null-space constrained policy optimization.arXiv preprint arXiv:2512.11391
5 Pith papers cite this work. Polarity classification is still indexing.
5
Pith papers citing it
citation-role summary
method 1
citation-polarity summary
years
2026 5verdicts
UNVERDICTED 5roles
method 1polarities
use method 1representative citing papers
MORA breaks the safety-helpfulness ceiling in LLMs by pre-sampling single-reward prompts and rewriting them to incorporate multi-dimensional intents, delivering 5-12.4% gains in sequential alignment and 4.6% overall improvement in simultaneous alignment.
Different LLM jailbreak techniques achieve similar harmful compliance but lead to distinct behavioral side effects and mechanistic changes.
PolicyAlign aligns LLMs to natural-language safety policies by synthesizing violating instructions and performing on-policy self-distillation with policy-sensitive filtering, improving safety without high-quality supervision data.
ShaPO improves LLM safety robustness over standard preference optimization by enforcing worst-case objectives via selective geometry control at token and reward levels.
citing papers explorer
-
Explaining and Breaking the Safety-Helpfulness Ceiling via Preference Dimensional Expansion
MORA breaks the safety-helpfulness ceiling in LLMs by pre-sampling single-reward prompts and rewriting them to incorporate multi-dimensional intents, delivering 5-12.4% gains in sequential alignment and 4.6% overall improvement in simultaneous alignment.