Attention Shift: Steering AI Away from Unsafe Content

This study investigates the generation of unsafe or harmful content in state-of-the-art generative models, focusing on methods for restricting such generations. We introduce a novel training-free approach using attention reweighing to remove unsafe concepts without additional training during inference. We compare our method against existing ablation methods, evaluating the performance on both, direct and adversarial jailbreak prompts, using qualitative and quantitative metrics. We hypothesize potential reasons for the observed results and discuss the limitations and broader implications of content restriction.

Unmasking the Veil: An Investigation into Concept Ablation for Privacy and Copyright Protection in Images

In this paper, we extend the study of concept ablation within pre-trained models as introduced in 'Ablating Concepts in Text-to-Image Diffusion Models' by (Kumari et al.,2022). Our work focuses on reproducing the results achieved by the different variants of concept ablation proposed and validated through predefined metrics. We also introduce a novel variant of concept ablation, namely 'trademark ablation'. This variant combines the principles of memorization and instance ablation to tackle the nuanced influence of proprietary or branded elements in model outputs. Further, our research contributions include an observational analysis of the model's limitations. Moreover, we investigate the model's behavior in response to ablation leakage-inducing prompts, which aim to indirectly ablate concepts, revealing insights into the model's resilience and adaptability. We also observe the model's performance degradation on images generated by concepts far from its target ablation concept, documented in the appendix.