List your company

Question Answering on TriviaQA (AUC, F95, AUP)

89.48AUCSIVR (Ours)
35.680849.647963.61577.5821Apr 17, 2026

Evaluation results

MethodLinks
SIVR (Ours)Backbone=Llama-3.1-8B-Instruct2026.0489.4839.8189.27
TADBackbone=Llama-3.1-8B-Instruct2026.0487.0250.5389.48
SAPLMABackbone=Llama-3.1-8B-Instruct2026.0483.762.3885.63
SARBackbone=Llama-3.1-8B-Instruct2026.0483.6645.2484.35
Lookback LensBackbone=Llama-3.1-8B-Instruct2026.0482.9459.0987.25
SATMD + MSPBackbone=Llama-3.1-8B-Instruct2026.0482.4561.3978.79
SEBackbone=Llama-3.1-8B-Instruct2026.0481.6554.7683.37
EntropyBackbone=Llama-3.1-8B-Instruct2026.0478.5867.0676.25
PerplexityBackbone=Llama-3.1-8B-Instruct2026.0478.1865.6875.39
Max ProbBackbone=Llama-3.1-8B-Instruct2026.0477.366.0873.94
Temp-ScaledBackbone=Llama-3.1-8B-Instruct2026.0476.6566.8672.71
EnergyBackbone=Llama-3.1-8B-Instruct2026.0467.587.1867.47
CoE-CBackbone=Llama-3.1-8B-Instruct2026.0465.1493.8968.58
P(True)Backbone=Llama-3.1-8B-Instruct2026.0465.0493.2965.35
CoE-RBackbone=Llama-3.1-8B-Instruct2026.0437.7510044.87