arXiv cs.AI· Anubhav Jain, Takashi Shibuya, Yuya Kobayashi, Yuhta Takida, Nasir Memon, Julian Togelius, Yuki Mitsufuji·· 12 小时前AI 评分31
TraSCE：无需训练修复负提示词在越狱提示下的引导崩溃
Rethinking Negative Prompting for Concept Suppression: Unconditional Anchoring and Localized Guidance
AI 导读
针对负提示词在越狱提示下出现的“引导崩溃”问题，研究者提出免训练推理时方法 TraSCE，将崩溃时的回退目标从负提示词预测改为中性无条件预测，并用对齐局部化引导步骤主动触发该回退。在面向裸露与暴力的对抗提示基准上，TraSCE 显著提升了负提示词对越狱提示的鲁棒性，且无需训练、权重更新或预先收集概念专属数据，对通用图像质量影响较小。该方法还可用于抑制艺术风格与特定物体。
正文 · 原文
Abstract:Text-to-image diffusion models have rapidly transitioned from research prototypes to widely deployed tools. Along with their impressive capabilities, these models can also generate harmful or undesirable content (e.g., NSFW imagery), motivating practical mechanisms for removing such concepts. Negative prompting, which uses the negative-prompt prediction as the guidance anchor, is among the most widely used mechanisms for steering sampling trajectories away from unwanted concepts. However, for jailbreak prompts, it can fail in a regime we refer to as \emph{guidance collapse}, where the model's denoising predictions for the prompt and negative prompt become similar, so that the guidance difference driving negative prompting vanishes and the guided prediction degenerates to the negative-prompt prediction, undermining suppression. To fix this issue, we propose TraSCE (Trajectory Steering for ConcEpt Suppression), a training-free inference-time method that makes negative prompting reliable under guidance collapse. TraSCE combines two complementary components: (i) unconditional anchoring, which makes the fallback under guidance collapse a neutral unconditional prediction rather than the negative-prompt prediction, and (ii) an alignment-localized steering step, which deliberately induces this collapse for adversarial prompts and steers the sample to the unconditional fallback, while remaining minimally intrusive on benign prompts. We evaluate TraSCE on adversarial prompt benchmarks targeting nudity and violence, and further demonstrate its applicability to suppressing artistic styles and objects. Across these settings, TraSCE substantially improves the robustness of negative prompting to jailbreak prompts without any training, weight updates, or pre-collected concept-specific data (prompts or images), while keeping the impact on general image quality small.
| Subjects: | Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Machine Learning (cs.LG) |
| Cite as: | arXiv:2412.07658 [cs.CV] |
| (or arXiv:2412.07658v3 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2412.07658 arXiv-issued DOI via DataCite |
Submission history
From: Anubhav Jain [view email]
[v1]
Tue, 10 Dec 2024 16:45:03 UTC (12,986 KB)
[v2]
Mon, 17 Mar 2025 15:37:35 UTC (29,569 KB)
[v3]
Fri, 16 Oct 2026 01:09:41 UTC (24,530 KB)
来源：arXiv cs.AI · arxiv.org