List your company

Classification on CSQA

2.21Expected Calibration Error (ECE)CaliDist (As.)
1.27687.575913.87520.1741Jun 4, 2026

Evaluation results

MethodLinks
CaliDist (As.)LLM=Phi-4-mini (3.8B)2026.062.2116.49
CaliDist (As.)LLM=Gemma-3 (4B)2026.062.3218.75
CaliDist (Sa.)LLM=Gemma-3 (4B)2026.062.5118.32
CaliDist (Pr.)LLM=Gemma-3 (4B)2026.062.5517.36
CaliDist (Pr.)LLM=Phi-4-mini (3.8B)2026.062.8916.62
CaliDist (Sa.)LLM=Phi-4-mini (3.8B)2026.063.3516.58
Temperature ScalingLLM=Phi-4-mini (3.8B)2026.064.5716.38
VanillaLLM=Phi-4-mini (3.8B)2026.0611.1317.73
ConsistencyLLM=Phi-4-mini (3.8B)2026.0611.5225.67
FSDLLM=Phi-4-mini (3.8B)2026.0614.6520.19
EntropyLLM=Phi-4-mini (3.8B)2026.0620.4630.34
Temperature ScalingLLM=Gemma-3 (4B)2026.0623.8724.18
VanillaLLM=Gemma-3 (4B)2026.0625.4425.43
EntropyLLM=Gemma-3 (4B)2026.0625.4825.53
FSDLLM=Gemma-3 (4B)2026.0625.4925.53
ConsistencyLLM=Gemma-3 (4B)2026.0625.5425.58