A binomial multibit watermarking scheme encodes every payload bit at each LLM token with dynamic redirection, outperforming baselines in accuracy and robustness for large payloads.
A watermark for large language models
9 Pith papers cite this work. Polarity classification is still indexing.
9
Pith papers citing it
citation-role summary
background 3
citation-polarity summary
verdicts
UNVERDICTED 9roles
background 3polarities
background 3representative citing papers
RLSpoofer trains a 4B model on 100 watermarked paraphrase pairs to spoof PF watermarks at 62% success rate, far exceeding baselines trained on up to 10,000 samples.
A learned transformation matrix minimizes CMI in teacher logits to degrade distillation performance while preserving task accuracy.
TAB-DRW embeds detectable watermarks in the frequency domain of normalized synthetic tabular data via DFT and rank-based pseudorandom bits, achieving robustness to attacks while preserving fidelity and supporting mixed data types.
Steer-to-Detect learns a steering vector injected into LLM hidden states to boost class separability and applies hypothesis testing with finite-sample Type I/II error guarantees for generated-text detection.
ArcMark is a multi-byte LLM watermark that achieves distortion-free embedding of several bytes per few hundred tokens by treating generation as a channel coding problem and using optimal transport to match distributions.
LLMs hallucinate citations at rates from 14.23% to 94.93%, with 1.07% of papers containing invalid citations and an 80.9% increase in 2025.
Derives matched converse and achievability bounds that characterize optimal trade-offs among false-alarm probability, detection error probability, distortion, and information rate for multi-bit watermarking of stationary ergodic stochastic processes.
LiSCP detects LLM-generated text via stylistic consistency profiling across paraphrased variants and reports up to 11.79% better cross-domain accuracy plus robustness to adversarial attacks.
citing papers explorer
-
Every Bit, Everywhere, All at Once: A Binomial Multibit LLM Watermark
A binomial multibit watermarking scheme encodes every payload bit at each LLM token with dynamic redirection, outperforming baselines in accuracy and robustness for large payloads.
-
RLSpoofer: A Lightweight Evaluator for LLM Watermark Spoofing Resilience
RLSpoofer trains a 4B model on 100 watermarked paraphrase pairs to spoof PF watermarks at 62% success rate, far exceeding baselines trained on up to 10,000 samples.
-
Towards Distillation-Resistant Large Language Models: An Information-Theoretic Perspective
A learned transformation matrix minimizes CMI in teacher logits to degrade distillation performance while preserving task accuracy.
-
Robust Spectral Watermark for Synthetic Tabular Data
TAB-DRW embeds detectable watermarks in the frequency domain of normalized synthetic tabular data via DFT and rank-based pseudorandom bits, achieving robustness to attacks while preserving fidelity and supporting mixed data types.
-
Steer-to-Detect: Probing Hidden Representations for Detection of LLM-Generated Texts
Steer-to-Detect learns a steering vector injected into LLM hidden states to boost class separability and applies hypothesis testing with finite-sample Type I/II error guarantees for generated-text detection.
-
ArcMark: Distortion-Free Multi-Byte LLM Watermark via Optimal Transport
ArcMark is a multi-byte LLM watermark that achieves distortion-free embedding of several bytes per few hundred tokens by treating generation as a channel coding problem and using optimal transport to match distributions.
-
GhostCite: A Large-Scale Analysis of Citation Validity in the Age of Large Language Models
LLMs hallucinate citations at rates from 14.23% to 94.93%, with 1.07% of papers containing invalid citations and an 80.9% increase in 2025.
-
Fundamental Trade-Offs in Multi-Bit Watermarking of Stochastic Processes
Derives matched converse and achievability bounds that characterize optimal trade-offs among false-alarm probability, detection error probability, distortion, and information rate for multi-bit watermarking of stationary ergodic stochastic processes.
-
Lightweight Stylistic Consistency Profiling: Robust Detection of LLM-Generated Textual Content for Multimedia Moderation
LiSCP detects LLM-generated text via stylistic consistency profiling across paraphrased variants and reports up to 11.79% better cross-domain accuracy plus robustness to adversarial attacks.