Across 51 quantized checkpoints, quality metrics fail to predict safety drops in 36 pairings and 10 hidden-danger cases, while a new RTSI screen routes all 10 dangerous rows to testing at matched bucket size.
Q-realign: Piggybacking realignment on quantization for safe and efficient LLM deployment.arXiv preprint arXiv:2601.08089, 2026
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
years
2026 3representative citing papers
The paper introduces a paired testing protocol for batch-conditioned refusal robustness in LLM serving and reports low rates of genuine safety-label flips after adjudication, with a batch-invariant kernel ablation eliminating observed flips.
Palette identifies refusal directions via multi-objective search, internalizes them through lightweight adaptation, and supports on-demand multi-domain authorization via independent learning and parameter merging.
citing papers explorer
-
Palette: A Modular, Controllable, and Efficient Framework for On-demand Authorized Safety Alignment Relaxation in LLMs
Palette identifies refusal directions via multi-objective search, internalizes them through lightweight adaptation, and supports on-demand multi-domain authorization via independent learning and parameter merging.