List your company

Question Answering on ConflictBank

80.2AccuracySABER
-318.640.261.8May 11, 2026

Evaluation results

MethodLinks
SABERBackbone=Llama-3.1-8B2026.0580.29788.192.5
SABERBackbone=Qwen-2.5-7B2026.0573.596.475.586
SABERBackbone=Qwen-2.5-3B2026.0571.697.673.385.4
SABERBackbone=Llama-3.2-3B2026.0570.997.682.389.9
TPCBackbone=Llama-3.1-8B2026.0569.692.336.464.3
Vanilla RAGBackbone=Llama-3.1-8B2026.0569.41001.750.9
Implicit-SCRBackbone=Qwen-2.5-7B2026.0568.693.113.353.2
Vanilla RAGBackbone=Qwen-2.5-7B2026.0566.5100050
TPCBackbone=Qwen-2.5-7B2026.0566.494.725.259.9
Implicit-SCRBackbone=Qwen-2.5-3B2026.0566.493.25.249.2
Vanilla RAGBackbone=Qwen-2.5-3B2026.0565.91000.950.4
TACS-LRBackbone=Qwen-2.5-7B2026.0565.890.713.352
Implicit-SCRBackbone=Llama-3.2-3B2026.0564.793.37.350.3
Vanilla RAGBackbone=Llama-3.2-3B2026.0563.51000.850.4
TPCBackbone=Qwen-2.5-3B2026.056390.924.157.5
CR-DPOBackbone=Qwen-2.5-7B2026.0562.885.741.363.5
TACS-LRBackbone=Qwen-2.5-3B2026.0562.389044.5
Internal-EvalBackbone=Llama-3.1-8B2026.0561.278.138.158.1
TACS-LRBackbone=Llama-3.2-3B2026.0560.790447
Context-EvalBackbone=Qwen-2.5-7B2026.0560.782.542.762.6
Internal-EvalBackbone=Llama-3.2-3B2026.0559.68162.171.5
Internal-EvalBackbone=Qwen-2.5-7B2026.0559.478.943.461.1
Context-EvalBackbone=Llama-3.1-8B2026.0558.961.690.376
Explicit-SCRBackbone=Qwen-2.5-7B2026.0558.681.228.754.9
TPCBackbone=Llama-3.2-3B2026.05578524.254.6
CR-DPOBackbone=Llama-3.1-8B2026.0552.350.385.868
Context-EvalBackbone=Llama-3.2-3B2026.0549.961.582.371.9
CR-DPOBackbone=Qwen-2.5-3B2026.0548.869.132.850.9
Explicit-SCRBackbone=Qwen-2.5-3B2026.0547.172.17.839.9
Internal-EvalBackbone=Qwen-2.5-3B2026.0541.853.551.752.6
Context-EvalBackbone=Qwen-2.5-3B2026.0539.647.370.759
CR-DPOBackbone=Llama-3.2-3B2026.0539.244.569.456.9
TACS-LRBackbone=Llama-3.1-8B2026.0533.846.83.725.2
Implicit-SCRBackbone=Llama-3.1-8B2026.0529.339.65.122.4
Explicit-SCRBackbone=Llama-3.2-3B2026.0527.833.25041.6
Closed-BookBackbone=Llama-3.1-8B2026.0523.50.610050.3
Explicit-SCRBackbone=Llama-3.1-8B2026.0522.223.828.726.3
Closed-BookBackbone=Qwen-2.5-7B2026.0518.81.110050.6
R-TuningBackbone=Qwen-2.5-7B2026.0518.55.15530.1
Closed-BookBackbone=Llama-3.2-3B2026.0516.80.310050.1
Closed-BookBackbone=Qwen-2.5-3B2026.0513.60.310050.1
R-TuningBackbone=Qwen-2.5-3B2026.058.61.940.921.4
R-TuningBackbone=Llama-3.1-8B2026.052.308.44.2
R-TuningBackbone=Llama-3.2-3B2026.050.200.80.4