DAPRO provides the first dynamic, theoretically guaranteed way to allocate interaction budgets across test cases for bounding time-to-event in multi-turn LLM evaluations, achieving tighter coverage than static conformal survival methods.
Conformalized survival analysis with adaptive cut-offs.Biometrika, 111(2):459–477
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.LG 2verdicts
UNVERDICTED 2representative citing papers
Introduces partial identification bounds and a double-robust SurvB-learner meta-learner for estimating robust CATE in survival analysis under informative censoring.
citing papers explorer
-
How Many Iterations to Jailbreak? Dynamic Budget Allocation for Multi-Turn LLM Evaluation
DAPRO provides the first dynamic, theoretically guaranteed way to allocate interaction budgets across test cases for bounding time-to-event in multi-turn LLM evaluations, achieving tighter coverage than static conformal survival methods.
-
Assessing the robustness of heterogeneous treatment effects in survival analysis under informative censoring
Introduces partial identification bounds and a double-robust SurvB-learner meta-learner for estimating robust CATE in survival analysis under informative censoring.