From the 1 of 8 linked papers with an AI index.
8 papers
Crossing the Margin Cliff: Toward Relearn-Robust LLM Unlearning via Margin Calibration
Xiangyu Yin, Jiaxu Liu, Zhen Chen +1
The paper identifies a systematic “margin cliff” that makes large language model unlearning vulnerable to relearn attacks and proposes Margin Calibration, a plug‑in method that add…
Cumulative Consensus Score: Label-Free and Model-Agnostic Evaluation of Object Detectors in Deployment
Avinaash Manoharan, Xiangyu Yin, Domenik Helm +1
Evaluating object detection models in deployment is challenging because ground-truth annotations are rarely available. We introduce the Cumulative Consensus Score (CCS), a label-fr…
Fragile by Design: On the Limits of Adversarial Defenses in Personalized Generation
Zhen Chen, Yi Zhang, Xiangyu Yin +4
Personalized AI applications such as DreamBooth enable the generation of customized content from user images, but also raise significant privacy concerns, particularly the risk of…
FALCON: Fine-grained Activation Manipulation by Contrastive Orthogonal Unalignment for Large Language Model
Jinwei Hu, Zhenglin Huang, Xiangyu Yin +4
Large language models have been widely applied, but can inadvertently encode sensitive or harmful information, raising significant safety concerns. Machine unlearning has emerged t…
TAIJI: Textual Anchoring for Immunizing Jailbreak Images in Vision Language Models
Xiangyu Yin, Yi Qi, Jinwei Hu +5
Vision Language Models (VLMs) have demonstrated impressive inference capabilities, but remain vulnerable to jailbreak attacks that can induce harmful or unethical responses. Existi…
CeTAD: Towards Certified Toxicity-Aware Distance in Vision Language Models
Xiangyu Yin, Jiaxu Liu, Zhen Chen +4
Recent advances in large vision-language models (VLMs) have demonstrated remarkable success across a wide range of visual understanding tasks. However, the robustness of these mode…