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.
Mitigating the safety alignment tax with null-space constrained policy optimization
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
citation-role summary
method 1
citation-polarity summary
years
2026 3verdicts
UNVERDICTED 3roles
method 1polarities
use method 1representative citing papers
Different LLM jailbreak techniques achieve similar harmful compliance but lead to distinct behavioral side effects and mechanistic changes.
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.
-
Different Paths to Harmful Compliance: Behavioral Side Effects and Mechanistic Divergence Across LLM Jailbreaks
Different LLM jailbreak techniques achieve similar harmful compliance but lead to distinct behavioral side effects and mechanistic changes.
-
Revisiting Robustness for LLM Safety Alignment via Selective Geometry Control
ShaPO improves LLM safety robustness over standard preference optimization by enforcing worst-case objectives via selective geometry control at token and reward levels.