5 papers
ScaleSweep: Accurate NVFP4 Post-Training Quantization of LLMs via Block Scale Initialization
Li Lin, Xiaojun Wan
NVFP4 is a recently introduced hardware-supported FP4 format that improves the fidelity of 4-bit quantization through fine-grained block scales. However, existing NVFP4 scale initi…
QuantileMark: A Message-Symmetric Multi-bit Watermark for LLMs
Junlin Zhu, Baizhou Huang, Xiaojun Wan
As large language models become standard backends for content generation, practical provenance increasingly requires multi-bit watermarking. In provider-internal deployments, a key…
Enhancing LLM Watermark Resilience Against Both Scrubbing and Spoofing Attacks
Huanming Shen, Baizhou Huang, Xiaojun Wan
Watermarking is a promising defense against the misuse of large language models (LLMs), yet it remains vulnerable to scrubbing and spoofing attacks. This vulnerability stems from a…
: A Black-Box Scrubbing Attack on LLM Watermarks
Baizhou Huang, Xiao Pu, Xiaojun Wan
Watermarking has emerged as a prominent technique for LLM-generated content detection by embedding imperceptible patterns. Despite supreme performance, its robustness against adver…
Style-Compress: An LLM-Based Prompt Compression Framework Considering Task-Specific Styles
Xiao Pu, Tianxing He, Xiaojun Wan
Prompt compression condenses contexts while maintaining their informativeness for different usage scenarios. It not only shortens the inference time and reduces computational costs…