Related papers: WorldCup Sampling for Multi-bit LLM Watermarking

We present a study to benchmark representative watermarking methods in cross-lingual settings. The current literature mainly focuses on the evaluation of watermarking methods for the English language. However, the literature for evaluating…

Large language models (LLMs) have witnessed a meteoric rise in popularity among the general public users over the past few months, facilitating diverse downstream tasks with human-level accuracy and proficiency. Prompts play an essential…

Benchmark contamination poses a significant challenge to the reliability of Large Language Models (LLMs) evaluations, as it is difficult to assert whether a model has been trained on a test set. We introduce a solution to this problem by…

Diffusion models have made substantial advances in recent years, enabling high-quality image synthesis; however, the widespread dissemination and reuse of their outputs have introduced new challenges in intellectual property protection and…

Large Language Model (LLM) watermarking embeds detectable signals into generated text for copyright protection, misuse prevention, and content detection. While prior studies evaluate robustness using watermark removal attacks, these methods…

While existing audio watermarking techniques have achieved strong robustness against traditional digital signal processing (DSP) attacks, they remain vulnerable to neural resynthesis. This occurs because modern neural audio codecs act as…

Identifying LLM-generated code through watermarking poses a challenge in preserving functional correctness. Previous methods rely on the assumption that watermarking high-entropy tokens effectively maintains output quality. Our analysis…

Text watermarks for large language models (LLMs) have been commonly used to identify the origins of machine-generated content, which is promising for assessing liability when combating deepfake or harmful content. While existing…

As open-source large language models (LLMs) like Llama3 become more capable, it is crucial to develop watermarking techniques to detect their potential misuse. Existing watermarking methods either add watermarks during LLM inference, which…

With the advent of large language models (LLMs), numerous software service providers (SSPs) are dedicated to developing LLMs customized for code generation tasks, such as CodeLlama and Copilot. However, these LLMs can be leveraged by…

We investigate the radioactivity of text generated by large language models (LLM), i.e. whether it is possible to detect that such synthetic input was used to train a subsequent LLM. Current methods like membership inference or active IP…

With the development of large models, watermarks are increasingly employed to assert copyright, verify authenticity, or monitor content distribution. As applications become more multimodal, the utility of watermarking techniques becomes…

Watermarking, the practice of embedding imperceptible information into media such as images, videos, audio, and text, is essential for intellectual property protection, content provenance and attribution. The growing complexity of digital…

Misuse of LLM-generated text can be curbed by watermarking techniques that embed implicit signals into the output. We propose a watermark that partitions the vocabulary at each decoding step into three sets (Green/Yellow/Red) with fixed…

Decoding strategies play a central role in shaping the reasoning ability of large language models (LLMs). Traditional methods such as greedy decoding and beam search often suffer from error propagation, while sampling-based approaches…

The proliferation of large language models for code (CodeLMs) and open-source contributions has heightened concerns over unauthorized use of source code datasets. While watermarking provides a viable protection mechanism by embedding…

Large language models (LLMs) are often equipped with multi-sample decoding strategies. An LLM implicitly defines an arithmetic code book, facilitating efficient and embarrassingly parallelizable \textbf{arithmetic sampling} to produce…

Detecting whether copyright holders' works were used in LLM pretraining is poised to be an important problem. This work proposes using data watermarks to enable principled detection with only black-box model access, provided that the…

We propose dgMARK, a decoding-guided watermarking method for discrete diffusion language models (dLLMs). Unlike autoregressive models, dLLMs can generate tokens in arbitrary order. While an ideal conditional predictor would be invariant to…

Image generation algorithms are increasingly integral to diverse aspects of human society, driven by their practical applications. However, insufficient oversight in artificial Intelligence generated content (AIGC) can facilitate the spread…