ProvenAI measures transparency in multi-hop QA via answer correctness, citation fidelity, and ablation-based document influence on HotpotQA, reporting 53.53% accuracy and 71.55% fidelity while identifying a citation-influence gap.
A call to action for a secure-by-design generative AI paradigm.arXiv preprint arXiv:2510.00451, 2025
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
ProvenAI: Provenance-Native Traces of Evidence in Generated Answers
ProvenAI measures transparency in multi-hop QA via answer correctness, citation fidelity, and ablation-based document influence on HotpotQA, reporting 53.53% accuracy and 71.55% fidelity while identifying a citation-influence gap.