Watermarking Conditional Text Generation for AI Detection: Unveiling Challenges and a Semantic-Aware Watermark Remedy
This addresses the challenge of balancing detection and quality in AI-generated text for applications like content moderation, though it is incremental as it builds on existing watermarking approaches.
The paper tackles the problem of watermarking AI-generated text for detection, revealing that existing methods significantly harm conditional text generation performance, and introduces a semantic-aware watermarking algorithm that improves performance across models like BART and Flan-T5 in tasks such as summarization and data-to-text generation while maintaining detection ability.
To mitigate potential risks associated with language models, recent AI detection research proposes incorporating watermarks into machine-generated text through random vocabulary restrictions and utilizing this information for detection. While these watermarks only induce a slight deterioration in perplexity, our empirical investigation reveals a significant detriment to the performance of conditional text generation. To address this issue, we introduce a simple yet effective semantic-aware watermarking algorithm that considers the characteristics of conditional text generation and the input context. Experimental results demonstrate that our proposed method yields substantial improvements across various text generation models, including BART and Flan-T5, in tasks such as summarization and data-to-text generation while maintaining detection ability.