GenoTrace: Inheritable Watermarks for Genome Foundation Model Distillation
Abstract
Can a genome model retain a detectable record of the synthetic sequences used to train it? We study watermark inheritance through distillation with GenoTrace, a codon-aware extension of green-list watermarking. Two token-level factors modulate the teacher's generation bias using codon position and organism-specific codon usage. The resulting sequences train a smaller student, whose outputs are audited without an active watermark processor. In a three-seed GenomeOcean-500M-to-100M experiment, the joint configuration achieves a mean audit score of 17.88 and 94.5% detection at a fixed threshold. It retains 49.0% detection after key-aware token substitution, compared with 0% for the available single-seed plain-watermark comparator, and 47.0% after combined mechanism-targeted nucleotide edits. Additional experiments establish inherited signal across five organism-conditioned datasets and teacher-student size ratios up to 40. Component ablations and computational sequence-quality assays reveal distinct operating points for detection strength and coding coverage. GenoTrace provides a practical token-level construction and an empirical account of how genomic structure shapes inherited watermark signals. The findings concern shared-tokenizer distillation and the tested editing procedures, with calibration and biological utility treated as separate evaluation requirements.
Related
- Coupling of oceanographic state to the dark proteome: a foundation for genome-informed marine productivity modeling
- Creator 1.0: Phylogeny-Guided Stagewise Pretraining for Plant Genome Foundation Models across 1,000 Species
- PlantGeneAnn: a high-accuracy ab initio plant gene annotation tool fine-tuned from a genome foundation model
- An adaptive transformer-based model with entropy-driven knowledge distillation for single-channel EEG REM sleep detection
- Improved vision transformer for sports image classification via transfer learning and knowledge distillation
BibTeX
@misc{yang2026genotrace,
title = {{GenoTrace: Inheritable Watermarks for Genome Foundation Model Distillation}},
author = {Guang Yang and Fengchen Liu},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2609.35881},
eprint = {2609.35881},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.35881},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - GenoTrace: Inheritable Watermarks for Genome Foundation Model Distillation
AU - Guang Yang
AU - Fengchen Liu
PY - 2026
DA - 2026/10/01
PB - arXiv
DO - 10.48550/arxiv.2609.35881
UR - https://arxiv.org/abs/2609.35881
AB - Can a genome model retain a detectable record of the synthetic sequences used to train it? We study watermark inheritance through distillation with GenoTrace, a codon-aware extension of green-list watermarking. Two token-level factors modulate the teacher's generation bias using codon position and organism-specific codon usage. The resulting sequences train a smaller student, whose outputs are audited without an active watermark processor. In a three-seed GenomeOcean-500M-to-100M experiment, the joint configuration achieves a mean audit score of 17.88 and 94.5% detection at a fixed threshold. It retains 49.0% detection after key-aware token substitution, compared with 0% for the available single-seed plain-watermark comparator, and 47.0% after combined mechanism-targeted nucleotide edits. Additional experiments establish inherited signal across five organism-conditioned datasets and teacher-student size ratios up to 40. Component ablations and computational sequence-quality assays reveal distinct operating points for detection strength and coding coverage. GenoTrace provides a practical token-level construction and an empirical account of how genomic structure shapes inherited watermark signals. The findings concern shared-tokenizer distillation and the tested editing procedures, with calibration and biological utility treated as separate evaluation requirements.
ER -