ProvenanceGuard

mentions 2 type Organization feed RSS

// recent coverage 2 mentions

13:07

2026-09-29

huggingface.co

ai-agents

Getting the Source Right, Not Just the Fact: Source-Aware Verification for MCP Agents

A team of researchers published ProvenanceGuard, a post-generation verification layer for Model Context Protocol (MCP) agents that checks whether each claim in an LLM answer is supported by the specif…

04:00

2026-07-28

machinebrief.com

artificial-intelligence

ProvenanceGuard: Source-Aware Factuality Verification for MCP-Based LLM Agents

Researchers introduce ProvenanceGuard, a source-aware verifier for LLM agents using the Model Context Protocol (MCP), achieving a block F1 of 0.802 and source accuracy of 0.858 on a 40-trace held-out …

// co-occurs with top 7 entities

Model Context Protocol 2 arXiv 2 RAGAS 1 MiniCheck 1 AlignScore 1 SummaC 1 Hugging Face 1

// topics top 6 topics

large language models 2 ai safety 2 ai research 2 artificial intelligence 1 ai agents 1 agent protocols 1