[EMNLP 2024 Demo] MarkLLM: An Open-Source Toolkit for LLM Watermarking
Source code of paper "RLCSD: Reinforcement Learning with Contrastive On-Policy Self-Distillation"