The Internet's Reading Group

paper

In-Context Watermarks for Large Language Models

Liu, Y., Zhao, X., Kruegel, C., Song, D., & Bu, Y. (2025)

arxiv:2505.16934

1 read

adversarial ai: 1ai: 1llm: 1watermarking: 1

readto readsave

readers, most recent

  1. ~gregcleo.com[read]llmaiadversarial aiwatermarking

    Note: Indirect prompt injection to get LLMs to watermark their own output.

    link