Token rankings from language models are unique and NP-hard to forge, providing the first polynomially unforgeable model signature.
zkLLM: Zero Knowledge Proofs for Large Language Models
7 Pith papers cite this work, alongside 5 external citations. Polarity classification is still indexing.
7
Pith papers citing it
5
external citations · external index
citation-role summary
background 1
citation-polarity summary
roles
background 1polarities
background 1representative citing papers
Routine behavior of 85 popular browser-extension crypto wallets leaks user addresses, enabling cross-site tracking, address clustering, and potential deanonymization for 65-82% of 35 million users.
AgentFlow builds a framework-agnostic Agent Dependency Graph from agent program source code to support static analyses such as BOM generation and prompt-to-tool risk detection, evaluated on 5,399 real programs across five frameworks.
The paper delivers a systematization of knowledge on AI agent-blockchain interactions via a bidirectional trust framework, an Agent-Blockchain Interaction Model, a five-dimensional evaluation lens, and nine identified open problems.
Alert suppression in blockchain systems can be made quadratically expensive (Θ(n²) in the number of alerters), and the paper gives constant-time and sequential protocols that achieve this bound.
zkComposer decomposes monolithic zkML proofs into parallel sub-proofs linked by shared boundary commitments, yielding up to 6.84× lower prover time on GPT-2 without new cryptographic primitives.
ZKMLOps is an MLOps framework that uses zero-knowledge proofs to generate verifiable cryptographic evidence of AI model compliance without revealing confidential information.
citing papers explorer
-
Token Rankings are Unforgeable Language Model Signatures
Token rankings from language models are unique and NP-hard to forge, providing the first polynomially unforgeable model signature.
-
The Masks We (Think We) Wear: Privacy Threats of Browser-Extension Wallets in the Web3 Ecosystem
Routine behavior of 85 popular browser-extension crypto wallets leaks user addresses, enabling cross-site tracking, address clustering, and potential deanonymization for 65-82% of 35 million users.
-
AgentFlow: Building Agent Dependency Graphs for Static Analysis of Agent Programs
AgentFlow builds a framework-agnostic Agent Dependency Graph from agent program source code to support static analyses such as BOM generation and prompt-to-tool risk detection, evaluated on 5,399 real programs across five frameworks.
-
Toward Web 4.0: Bidirectional Trust between AI Agents and Blockchain
The paper delivers a systematization of knowledge on AI agent-blockchain interactions via a bidirectional trust framework, an Agent-Blockchain Interaction Model, a five-dimensional evaluation lens, and nine identified open problems.
-
Resilient Alerting Protocols for Blockchains
Alert suppression in blockchain systems can be made quadratically expensive (Θ(n²) in the number of alerters), and the paper gives constant-time and sequential protocols that achieve this bound.
-
zkComposer: Decomposing Proof Construction to Scale zkML
zkComposer decomposes monolithic zkML proofs into parallel sub-proofs linked by shared boundary commitments, yielding up to 6.84× lower prover time on GPT-2 without new cryptographic primitives.
-
"Show Me You Comply... Without Showing Me Anything": Zero-Knowledge Software Auditing for AI-Enabled Systems
ZKMLOps is an MLOps framework that uses zero-knowledge proofs to generate verifiable cryptographic evidence of AI model compliance without revealing confidential information.