MiRD decomposes overall miscoverage into sampling and conditional selection risks for conformal set-valued prediction in open-ended QA, bounding each while using the full calibration set.
arXiv preprint arXiv:2510.17897 , year=
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.CL 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
BalanceRAG uses sequential graphical testing on a 2D lattice of threshold pairs to certify safe operating points that meet target risk levels in cascaded RAG while increasing coverage.
citing papers explorer
-
MiRD: Reliable Set-Valued Prediction for Open-Ended Question Answering via Miscoverage Risk Decomposition
MiRD decomposes overall miscoverage into sampling and conditional selection risks for conformal set-valued prediction in open-ended QA, bounding each while using the full calibration set.
-
BalanceRAG: Joint Risk Calibration for Cascaded Retrieval-Augmented Generation
BalanceRAG uses sequential graphical testing on a 2D lattice of threshold pairs to certify safe operating points that meet target risk levels in cascaded RAG while increasing coverage.