AI agents can conduct undetectable covert conversations using a new pseudorandom noise-resilient key exchange that works without shared keys and with only constant min-entropy in messages.
Undetectable watermarks for language models
8 Pith papers cite this work, alongside 12 external citations. Polarity classification is still indexing.
8
Pith papers citing it
12
external citations · external index
citation-role summary
background 1
citation-polarity summary
roles
background 1polarities
background 1representative citing papers
Signature filtering learns unreliable tokens with MILP and removes them at detection time, raising true positive rates from 8-31% to 78-99% across Kgw, Sweet, Unigram, and Exp watermarks on multiple corpora and LLMs while controlling false positives.
Tool-using LLM agents can implement undetectable stegosystems, shifting the primary barrier to covert multi-agent collusion from technical feasibility to coordination without explicit agreement.
The Generalized Turing Test defines relative intelligence as the inability of one agent to distinguish an imitator from the original through interaction.
Response times modeled as drift-diffusion processes enable consistent estimation of population-average preferences from heterogeneous anonymous binary choices.
Watermark-based dataset inference achieves membership detection performance comparable to loss-based methods when subset exposure is high, under alternate assumptions.
LLM watermarking adoption is limited by misaligned stakeholder incentives; incentive-aligned approaches such as in-context watermarking can enable practical use in targeted domains like education and peer review.
The paper introduces a taxonomy of AI safety for LLMs organized into Trustworthy AI, Responsible AI, and Safe AI perspectives, accompanied by a review of state-of-the-art methods, challenges, and future directions.
citing papers explorer
-
Undetectable Conversations Between AI Agents via Pseudorandom Noise-Resilient Key Exchange
AI agents can conduct undetectable covert conversations using a new pseudorandom noise-resilient key exchange that works without shared keys and with only constant min-entropy in messages.
-
Signature filtering: a lightweight enhancement for statistical watermark detection in large language models
Signature filtering learns unreliable tokens with MILP and removes them at detection time, raising true positive rates from 8-31% to 78-99% across Kgw, Sweet, Unigram, and Exp watermarks on multiple corpora and LLMs while controlling false positives.
-
Tool Use Enables Undetectable Steganography in Multi-Agent LLM Systems
Tool-using LLM agents can implement undetectable stegosystems, shifting the primary barrier to covert multi-agent collusion from technical feasibility to coordination without explicit agreement.
-
The Generalized Turing Test: A Foundation for Comparing Intelligence
The Generalized Turing Test defines relative intelligence as the inability of one agent to distinguish an imitator from the original through interaction.
-
Response Time Enhances Alignment with Heterogeneous Preferences
Response times modeled as drift-diffusion processes enable consistent estimation of population-average preferences from heterogeneous anonymous binary choices.
-
Watermarking for Proprietary Dataset Protection
Watermark-based dataset inference achieves membership detection performance comparable to loss-based methods when subset exposure is high, under alternate assumptions.
-
Position: LLM Watermarking Should Align Stakeholders' Incentives for Practical Adoption
LLM watermarking adoption is limited by misaligned stakeholder incentives; incentive-aligned approaches such as in-context watermarking can enable practical use in targeted domains like education and peer review.
-
AI Safety Landscape for Large Language Models: Taxonomy, State-of-the-art, and Future Directions
The paper introduces a taxonomy of AI safety for LLMs organized into Trustworthy AI, Responsible AI, and Safe AI perspectives, accompanied by a review of state-of-the-art methods, challenges, and future directions.