List your company

Question Answering on ARC Easy (Accuracy and CC Metrics)

92.18Accuracy (ARC-E)Best Model
78.753682.239385.72589.2107Jun 9, 2025Jul 2, 2025Jul 25, 2025Aug 17, 2025Sep 9, 2025Oct 2, 2025Oct 26, 2025

Evaluation results

MethodLinks
Best ModelBackbone=LLaMA 3.1 8B, Shots=5-shot, #D=42025.1092.18————1.14—
BSBABackbone=LLaMA 3.1 8B, Shots=5-shot, #D=82025.1090.91————1.37—
BaselineBackbone=LLaMA 3.1 8B, Shots=5-shot2025.1090.36——————
LLaMA3.1-8BModel=LLaMA3.1-8B, Attribution Method (LIME/LIG)=LIME, Evaluation Mode (B-B, B-T, T-T)=B-B2025.0687.7213.3214.1715.03-1.912.0125.72
LLaMA3.1-8BModel=LLaMA3.1-8B, Attribution Method (LIME/LIG)=LIME, Evaluation Mode (B-B, B-T, T-T)=B-T2025.0687.7213.4714.3115.151.8415.0128.38
LLaMA3.1-8BModel=LLaMA3.1-8B, Attribution Method (LIME/LIG)=LIME, Evaluation Mode (B-B, B-T, T-T)=T-T2025.0686.7614.9715.8916.821.7614.8827.86
LLaMA3.1-8BModel=LLaMA3.1-8B, Attribution Method (LIME/LIG)=LIG, Evaluation Mode (B-B, B-T, T-T)=T-T2025.0684.841.541.822.0839.9347.0953.79
LLaMA3.1-8BModel=LLaMA3.1-8B, Attribution Method (LIME/LIG)=LIG, Evaluation Mode (B-B, B-T, T-T)=B-B2025.0684.260.981.351.6837.445.5552.88
LLaMA3.1-8BModel=LLaMA3.1-8B, Attribution Method (LIME/LIG)=LIG, Evaluation Mode (B-B, B-T, T-T)=B-T2025.0684.261.561.842.1138.9446.7653.63
LLaMA3.2-3BModel=LLaMA3.2-3B, Attribution Method (LIME/LIG)=LIME, Evaluation Mode (B-B, B-T, T-T)=B-B2025.0681.625.586.136.687.9720.7233.14
LLaMA3.2-3BModel=LLaMA3.2-3B, Attribution Method (LIME/LIG)=LIME, Evaluation Mode (B-B, B-T, T-T)=B-T2025.0681.625.586.116.668.9921.5334.19
LLaMA3.2-3BModel=LLaMA3.2-3B, Attribution Method (LIME/LIG)=LIME, Evaluation Mode (B-B, B-T, T-T)=T-T2025.0681.046.326.877.4211.4823.1634.8
LLaMA3.2-3BModel=LLaMA3.2-3B, Attribution Method (LIME/LIG)=LIG, Evaluation Mode (B-B, B-T, T-T)=B-B2025.0680.231.21.622.0133.7743.1551.75
LLaMA3.2-3BModel=LLaMA3.2-3B, Attribution Method (LIME/LIG)=LIG, Evaluation Mode (B-B, B-T, T-T)=B-T2025.0680.231.251.662.0334.8844.0252.52
LLaMA3.2-3BModel=LLaMA3.2-3B, Attribution Method (LIME/LIG)=LIG, Evaluation Mode (B-B, B-T, T-T)=T-T2025.0679.271.712.072.4337.9746.2453.74