List your company

Question Answering on CaseHOLD (eval)

7.8RiskLTT
7.310.67514.0517.425May 18, 2026

Evaluation results

MethodLinks
LTTN (EVAL size)=2,880, Err (Base verifier-fail rate)=34.0%, alpha* (Target risk level)=0.252026.057.8——
ConfFactN (EVAL size)=2,880, Err (Base verifier-fail rate)=34.0%, alpha* (Target risk level)=0.252026.0511.3——
CSAN (EVAL size)=2,880, Err (Base verifier-fail rate)=34.0%, alpha* (Target risk level)=0.252026.0520.349.80
5 online baselinesN (EVAL size)=2,880, Err (Base verifier-fail rate)=34.0%, alpha* (Target risk level)=0.252026.05——100
NEX-ConfN (EVAL size)=2,880, Err (Base verifier-fail rate)=34.0%, alpha* (Target risk level)=0.252026.05——0