Research
PoLO: Proof-of-Learning and Proof-of-Ownership at Once with Chained Watermarking
arXiv:2505.12296v2 Announce Type: replace-cross Abstract: Our evaluation shows that PoLO achieves extbf{99%} watermark detection accuracy for ownership verification, while preserving data privacy and
arXiv:2505.12296v2 Announce Type: replace-cross Abstract: Our evaluation shows that PoLO achieves extbf{99%} watermark detection accuracy for ownership verification, while preserving data privacy and cutting verification costs to just extbf{1.5--10%} of traditional methods. Forging PoLO demands extbf{1.1--4imes} more resources than honest proof generation, with the original proof retaining over extbf{90%} detection accuracy even after attacks.
Related
- Private Seeds, Public LLMs: Realistic and Privacy-Preserving Synthetic Data Generation
- PrivaDE: Privacy-preserving Data Evaluation for Blockchain-based Data Marketplaces
- A2-DIDM: Privacy-preserving Accumulator-enabled Auditing for Distributed Identity of DNN Model
- AdaProb: Efficient Machine Unlearning via Adaptive Probability
Source: arXiv cs.AI | 2026-04-27