ShaPO improves LLM safety robustness over standard preference optimization by enforcing worst-case objectives via selective geometry control at token and reward levels.
Alignmerge-alignment-preserving large language model merging via fisher-guided geometric constraints.arXiv preprint arXiv:2512.16245,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
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.