arXiv AI
Jun 2
Linguistics-Aware Non-Distortionary LLM Watermarking
arXiv:2606. 00613v1 Announce Type: cross Abstract: Watermarking should identify language-model output without degrading quality or limiting verification to the model provider.
TextSeal2: Distortion-Free LLM Watermarking with Copula Mixing and Memoryless Races
TextSeal2 is a distortion‑free watermarking method for large language models that maintains signal integrity during speculative decoding. It merges multiple keys into a single virtual key using a Gaussian copula, ensuring that the acceptance rate and output diversity remain unchanged while adding no measurable inference latency. Experiments on three LLMs show that TextSeal2 detects 8–22 percentage points more 100‑token answers than its predecessor TextSeal and outperforms textGrain under speculative decoding, all without post‑logits optimization.
Related stories
arXiv Machine Learning
Jun 18
Signature filtering: a lightweight enhancement for statistical watermark detection in large language models
arXiv:2606. 18430v1 Announce Type: new Abstract: Statistical watermarks help organizations attribute large language model (LLM) outputs, yet existing detectors often struggle when watermark signals are weak, texts are repetitive, or watermarks are edited.
arXiv AI
Oct 21
A Unified Attack Framework against Statistical Generative Watermarks for Large Language Models
arXiv:2610.27653v1 Announce Type: cross Abstract: Watermarking has emerged as a key tool for identifying text generated by large language models (LLMs), with current schemes embedding statistical sig...
arXiv Computation and Language
Oct 15
Majority Bit-Aware Watermarking For Large Language Models
arXiv:2508.03829v3 Announce Type: replace Abstract: The growing deployment of Large Language Models (LLMs) has raised concerns about their misuse in generating harmful or deceptive content. To addres...
arXiv Machine Learning
Jul 7
The Truncation Blind Spot: How Decoding Strategies Systematically Exclude Human-Like Token Choices
arXiv:2603. 18482v2 Announce Type: replace-cross Abstract: Standard decoding strategies for text generation, including top-$k$, nucleus sampling, and contrastive search, select tokens based on likelihood, restricting outputs to high-probability regions.
arXiv AI
Aug 5
Oilbird: Training-Free Speculative Decoding with Keys the Verifier Already Computes
arXiv:2608. 03839v1 Announce Type: new Abstract: Training-free speculative decoding drafts by matching an exact suffix of the context against a pool of earlier context.
arXiv AI
Oct 31
Only Accepted Tokens Count: An Energy Accounting of Speculative Decoding Under Latency Targets
The paper investigates how speculative decoding—where a draft model proposes multiple future tokens that are then verified by the target model—affects energy consumption in large language model serving under latency constraints. It finds that energy savings depend heavily on the acceptance rate of drafted tokens, which varies by model architecture, drafting scheme, and the type of text (e.g., conversation vs. code, English vs. Romansh). The study shows that while speculation can reduce energy per token at small batch sizes, at larger batches the verification cost dominates, potentially turning speculation into an energy loss, especially when acceptance rates are low.
arXiv AI
Oct 9
Speedbumps: Rejection Attacks on Speculative Decoding
Speedbumps: Rejection Attacks on Speculative Decoding explores how Speculative Rejection Attacks (SRAs) can degrade the efficiency of speculative decoding by causing draft and target models to disagree more often, leading to more target-model forward passes per token. The authors introduce two attacks—Speedbump-P and Speedbump-D—that append adversarial suffixes to attacker-controlled content, optimizing the expected length of the accepted speculative prefix. In some scenarios, these attacks slow inference below autoregressive decoding, reduce output quality, yet remain effective under sampling and transfer across drafters or target models sharing a drafter.
arXiv Machine Learning
Oct 23
MarkTune: Improving the Quality-Detectability Trade-off in Model-Embedded LLM Watermarking
MarkTune is a model‑embedded watermarking framework that improves the trade‑off between generation quality and detectability. By regularizing toward a high‑quality reference distribution, it decouples quality preservation from proximity to the base model and uses the GaussMark detection statistic as a reward during on‑policy fine‑tuning. Empirical results show that MarkTune enhances the quality‑detectability frontier, maintains downstream task performance, and remains robust against text‑editing and open‑weight attacks across diverse datasets.
arXiv Machine Learning
Jul 30
Adaptively Robust LLM Monitoring via Activation Watermarking
arXiv:2603. 23171v3 Announce Type: replace-cross Abstract: Providers monitor deployed large language models (LLMs) to detect misuse that they cannot prevent.
arXiv Machine Learning
Jul 14
One Token Is Enough: Fingerprinting and Verifying Large Language Models from Single-Token Output Distributions
arXiv:2607. 10252v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly consumed through opaque serving chains - API aggregators, resellers, and inference providers - in which the client has no technical means to confirm that the model answering is the model advertised, and recent audits show that a substantial fraction of commercial endpoints deviate from the vendor's reference weights.
arXiv Computation and Language
Oct 19
Watermarkability: Measuring and Improving Language-Model Compatibility with Logit-Based Watermarks
The paper introduces "watermarkability," a metric that quantifies how compatible a language model is with a logit‑based watermark that boosts the probability of a specified token set during decoding. By measuring the expected minimum KL divergence needed to achieve a target probability mass, the authors can assess and compare models and watermark rules. They propose a two‑stage approach—first training a low‑rank adapter conditioned on the target set, then applying a minimal uniform logit bias—to reduce this cost, achieving significant perplexity reductions across multiple watermark schedules on OPT‑125M and Qwen3‑8B.