AINeutralarXiv – CS AI · Mar 266/10
🧠
Is Multilingual LLM Watermarking Truly Multilingual? Scaling Robustness to 100+ Languages via Back-Translation
Researchers demonstrate that current multilingual watermarking methods for LLMs fail to maintain robustness across medium- and low-resource languages, particularly under translation attacks. They introduce STEAM, a new detection method using Bayesian optimization that improves watermark detection across 133 languages with significant performance gains.