paperAugust 2026Unreviewed

Semantic-Oriented Robust Sentence Level Text Watermark for Large Language Models

Bo Li, Kun Zhang, Chengyou Song, Xi Chen, Hailin Zhou, Richang Hong

IEEE Transactions on Computational Social Systems

Abstract

Text watermarking focuses on embedding identifiable information into the generated content, which has become increasingly important with the rapid development of large language models (LLMs). Existing watermarking works either divide the vocabulary of LLMs into “green” and “red” tokens for the watermark generation (i.e., token-level watermark), or use the distance of generated sentence embeddings to distinguish the “green” and “red” partitions (i.e., sentence-level watermark). Despite the achiev

Cite

@article{li2026semanticoriented,
  title = {{Semantic-Oriented Robust Sentence Level Text Watermark for Large Language Models}},
  author = {Bo Li and Kun Zhang and Chengyou Song and Xi Chen and Hailin Zhou and Richang Hong},
  year = {2026},
  month = aug,
  journal = {IEEE Transactions on Computational Social Systems},
  doi = {10.1109/TCSS.2026.3687267},
  url = {https://www.semanticscholar.org/paper/de0e56e50fc26303b1d9b0a7362e5a2eda32334a}
}