List your company

Calibration on HotpotQA

0.28ECENAACL
0.267560.351530.43550.51947Jan 16, 2026

Evaluation results

MethodLinks
NAACLBackbone=Llama-3.1-8B-Instruct2026.010.280.778
Noise-awareBackbone=Llama-3.1-8B-Instruct2026.010.3090.642
NAACLBackbone=Qwen2.5-7B-Instruct2026.010.3120.735
Label-only SFTBackbone=Llama-3.1-8B-Instruct2026.010.3190.711
Label-only SFTBackbone=Qwen2.5-7B-Instruct2026.010.3210.699
Noise-awareBackbone=Qwen2.5-7B-Instruct2026.010.3250.692
CoTBackbone=Qwen2.5-7B-Instruct2026.010.3530.693
NAACLBackbone=DeepSeek-R1-Distill-Llama-8B2026.010.3590.663
EnsembleBackbone=Qwen2.5-7B-Instruct2026.010.380.681
VanillaBackbone=Qwen2.5-7B-Instruct2026.010.3910.712
NAACLBackbone=DeepSeek-R1-Distill-Qwen-7B2026.010.3910.702
EnsembleBackbone=Llama-3.1-8B-Instruct2026.010.3970.65
Noise-awareBackbone=DeepSeek-R1-Distill-Qwen-7B2026.010.4220.523
Noise-awareBackbone=DeepSeek-R1-Distill-Llama-8B2026.010.4430.633
CoTBackbone=Llama-3.1-8B-Instruct2026.010.4440.645
VanillaBackbone=DeepSeek-R1-Distill-Llama-8B2026.010.4570.66
VanillaBackbone=Llama-3.1-8B-Instruct2026.010.460.605
EnsembleBackbone=DeepSeek-R1-Distill-Llama-8B2026.010.4650.65
CoTBackbone=DeepSeek-R1-Distill-Llama-8B2026.010.4840.617
VanillaBackbone=DeepSeek-R1-Distill-Qwen-7B2026.010.4990.641
EnsembleBackbone=DeepSeek-R1-Distill-Qwen-7B2026.010.5150.614
Label-only SFTBackbone=DeepSeek-R1-Distill-Llama-8B2026.010.5170.577
CoTBackbone=DeepSeek-R1-Distill-Qwen-7B2026.010.5290.632
Label-only SFTBackbone=DeepSeek-R1-Distill-Qwen-7B2026.010.5910.538