“Related to this idea, there are interesting papers that explicitly examine the adversarial case. Basically, besides the provider hiding watermarks, one could also think of an adversary training a model to exhibit this behaviour depending on the Input of the prompt. So if you use the manipulated model, not only information about the author that the platfor…”
TrojanStego: Your Language Model Can Secretly Be A Steganographic Privacy Leaking Agent
As large language models (LLMs) become integrated into sensitive workflows, concerns grow over their potential to leak confidential information. We propose TrojanStego, a novel threat model in which an adversary fine-tunes an LLM to embed sensitive context information into natural-looking outputs via linguistic steganography, without requiring explicit control over inference inputs. We introduce a taxonomy outlining risk factors for compromised LLMs, and use it to evaluate the risk profile of the threat. To implement TrojanStego, we propose a practical encoding scheme based on vocabulary partitioning learnable by LLMs via fine-tuning. Experimental results show that compromised models reliably transmit 32-bit secrets with 87% accuracy on held-out prompts, reaching over 97% accuracy using majority voting across three generations. Further, they maintain high utility, can evade human detection, and preserve coherence. These results highlight a new class of LLM data exfiltration attacks that are passive, covert, practical, and dangerous.
候选层:至少 2 个不同的人、在 2 个不同月份提到过,还没到主书架的 3 人 3 个月。没有做立场标注,所以只写“被提及”。
HN 书评
推荐在前,同类按时间 · 每篇最多保留 20 条,统计用全部提及- TrojanStego: Your Language Model Can Be a Steganographic Agent ↗Worta · 2025-05-31 · 1 分
- Language Model Can Be a Steganographic Privacy Leaking Agent ↗dennis-tra · 2025-07-29 · 3 分