Symmetry Defeats Auditing
Not on the map. Why?
Abstract (arXiv)
We demonstrate an attack on Introspection Adapters (Shenoy et al., 2026).
Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 8 of its 10 references that are in the map. First posted on arXiv 27 May 2026.
Builds on
8 of 10 references.
References · 8
- 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,380
- 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,415
- 2024Safe LoRA: the Silver Lining of Reducing Safety Risks when Fine-tuning Large Language Modelscited by 125
- 2024LoRA vs Full Fine-tuning: An Illusion of Equivalencecited by 109
- 2024Towards Understanding Safety Alignment: A Mechanistic Perspective from Safety Neuronscited by 53
- 2025PLoP: Precise LoRA Placement for Efficient Finetuning of Large Modelscited by 14
- 2026Introspection Adapters: Training LLMs to Report Their Learned Behaviorscited by 11
- 2025Safe Pruning LoRA: Robust Distance-Guided Pruning for Safety Alignment in Adaptation of LLMscited by 10
Built on by
Not cited yet.