3 papers
cs.IT2026
Pair-Dependent Judge Errors Change the Minimax Rate of Semantic Entropy
Jiamiao Liu, Dewen Qiao, Yu Zhang +1
Semantic entropy, the entropy of the meanings among a language model's sampled answers, is estimated from the verdicts of a judge that declares pairs of answers equivalent or not,…
cs.LG2026
Certified Against Which Oracle? Execution Labels Set the Reported Risk of Conformal Abstention for Text-to-SQL
Jiamiao Liu, Dewen Qiao, Yu Zhang +1
A conformal abstention certificate for text-to-SQL is only as truthful as the correctness labels it is calibrated on. The uncertainty pipelines that read confidence off execution c…
cs.CL2026
Certify or Refuse: A Cross-Model Map for Selective Risk Control with Coverage Floors under Covariate Shift
Jiamiao Liu, Dewen Qiao, Yu Zhang +1
Certified selective predictors attain whatever coverage they attain; operators impose an automation floor: answer at least a -fraction of shifted target traffic with at most an…