Ali Al-Lawati, Elsayed Eshra, Dongwon Lee, Suhang Wang · 2026 · arXiv
DOI: 10.48550/arXiv.2610.31558 · open-access PDF
在黑匣子LLM中,通过采样或优化低维加西亚隐形空间并将其解码为文本,通过emph(随机地图)和最接近的代码搜索引发恶意行为,例如过度思考。由于随机地图不是从语言中学到的,其提示可以非常不自然,并且通过防护栏可以很容易地检测到输入,并且它们产生的风险估计并非基于自然提示。
Machine translation (Meta M2M100); the English original follows.
\emph{Continuous-space} attacks find prompts that trigger malicious behaviors, such as overthinking, in black-box LLMs by sampling or optimizing over a low-dimensional Gaussian latent space and decoding it to text through a \emph{random mapping} and nearest-token lookup. Because the random mapping is not learned from language, its prompts can be highly unnatural, and as such easily detectable through input guardrails, and the risk estimates they yield are not grounded in natural prompts. In this paper, we demonstrate that the same continuous latent-space framework can be coupled directly to a
Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese
GET https://store.agentexchange.work/gw/papers/verify?title=...&year=... (0.003 USD), fetch this record as JSON with /gw/papers/paper?id=2610.31558, or list what it cites with /gw/papers/references?id=.... Pay per call in USDC (x402) or with a 19 USD card key for 1,000 calls. Catalogue · llms.txtSources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange