Selective Disclosure Watermarking for Large Language Models
Watermarking methods embed imperceptible and verifiable signals into text generated by large language models (LLMs). Existing approaches include zero-bit schemes for distinguishing synthetic text from human writing and multi-bit schemes for embedding metadata. However, current multi-bit watermarking methods do not allow selective disclosure: verifying any part of the watermark requires revealing the entire embedded message. This lack of control leads to unnecessary information exposure and raises privacy concerns. We propose Hierarchical Vocabulary Routing (HeRo), a watermarking framework that
Lineage graph
Paper → model → repo connections mined from source citations (Tier-1 exact match).
Why these links exist
Every edge carries a method, confidence, and the source snippet that justified it — so bad links are debuggable.
- PossiblePossibly related (embedding) · 45%Coders Say They Already Found Workarounds to Claude’s Invisible Watermarks →
- PossiblePossibly related (embedding) · 53%Implementing Watermarking for Language Models [P] →
- LinkedLinked via arxiv author · 85%Xuyang Chen →
“Selective Disclosure Watermarking for Large Language Models”
- LinkedLinked via arxiv author · 85%Xiang Li →
“Selective Disclosure Watermarking for Large Language Models”
- LinkedLinked via arxiv author · 85%Yangxinyu Xie →
“Selective Disclosure Watermarking for Large Language Models”
- LinkedLinked via arxiv author · 85%Qi Long →
“Selective Disclosure Watermarking for Large Language Models”
- PossiblePossibly related (embedding) · 47%Anthropic says text watermarking scheme relies on inconsequential words →
- PossiblePossibly related (embedding) · 50%Anthropic explains how Claude’s invisible text watermarks will work →