THU-KEG/WaterBench
[ACL2024-Main] Data and Code for WaterBench: Towards Holistic Evaluation of LLM Watermarks
This project helps developers and researchers evaluate the effectiveness of different watermarking techniques for Large Language Models (LLMs). It takes various LLM outputs, applies different watermark algorithms, and provides metrics such as detection z-scores and GPT-4 based evaluation results. The primary users are researchers or engineers working on LLM security, content provenance, or responsible AI.
No commits in the last 6 months.
Use this if you are a machine learning researcher or engineer who needs to systematically test and compare how well different watermarks perform on LLMs across various datasets and models.
Not ideal if you are looking for a plug-and-play solution to apply watermarks to your LLMs without needing to dive into evaluation metrics or experiment with different watermark parameters.
Stars
30
Forks
2
Language
Python
License
MIT
Category
Last pushed
Nov 14, 2023
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/transformers/THU-KEG/WaterBench"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
THU-BPM/MarkLLM
MarkLLM: An Open-Source Toolkit for LLM Watermarking.（EMNLP 2024 System Demonstration)
git-disl/Vaccine
This is the official code for the paper "Vaccine: Perturbation-aware Alignment for Large...
zjunlp/Deco
[ICLR 2025] MLLM can see? Dynamic Correction Decoding for Hallucination Mitigation
HillZhang1999/ICD
Code & Data for our Paper "Alleviating Hallucinations of Large Language Models through Induced...
voidism/DoLa
Official implementation for the paper "DoLa: Decoding by Contrasting Layers Improves Factuality...