List your company

Natural Language Inference on MNLI (ECE, BS)

2.74ECECaliDist (As.)
1.5649.50217.4425.378Jun 4, 2026

Evaluation results

MethodLinks
CaliDist (As.)LLM=LLAMA-3.1 8B2026.062.7419.01
CaliDist (Sa.)LLM=LLAMA-3.1 8B2026.062.7719.01
VanillaLLM=LLAMA-3.1 8B2026.062.81—
CaliDist (Pr.)LLM=LLAMA-3.1 8B2026.062.8319.02
Temperature ScalingLLM=LLAMA-3.1 8B2026.065.8319.27
CaliDist (Sa.)LLM=QWEN-3 8B2026.066.3821.43
CaliDist (As.)LLM=QWEN-3 8B2026.066.4921.36
CaliDist_verbalized (Pr.)LLM=LLAMA-3.1 8B2026.06722.37
CaliDist_verbalized (Sa.)LLM=QWEN-2 7B2026.067.2718.98
CaliDist_verbalized (As.)LLM=QWEN-2 7B2026.067.7919.05
CaliDist_verbalized (Pr.)LLM=QWEN-2 7B2026.069.1819.1
VerbalizedLLM=QWEN-2 7B2026.069.2819.21
CaliDist_verbalized (As.)LLM=LLAMA-3.1 8B2026.0610.0323.22
CaliDist_verbalized (Sa.)LLM=LLAMA-3.1 8B2026.0610.223.26
CaliDist (Pr.)LLM=QWEN-3 8B2026.0611.0422
ConsistencyLLM=LLAMA-3.1 8B2026.0611.2921.03
CaliDist_verbalized (Sa.)LLM=GEMMA-3 4B2026.0611.9825.72
CaliDist_verbalized (As.)LLM=GEMMA-3 4B2026.0617.8627.92
FSDLLM=LLAMA-3.1 8B2026.0617.9924.51
CaliDist_verbalized (Pr.)LLM=GEMMA-3 4B2026.0619.2629.31
CaliDist_verbalized (Sa.)LLM=PHI-4-MINI 3.8B2026.0619.7724.63
CaliDist_verbalized (Pr.)LLM=PHI-4-MINI 3.8B2026.0619.9127.08
CaliDist_verbalized (As.)LLM=PHI-4-MINI 3.8B2026.0622.4426.09
VerbalizedLLM=PHI-4-MINI 3.8B2026.0624.7324.98
VerbalizedLLM=LLAMA-3.1 8B2026.0626.428.49
VerbalizedLLM=GEMMA-3 4B2026.0626.530.42
Temperature ScalingLLM=QWEN-3 8B2026.0627.1628.35
VanillaLLM=QWEN-3 8B2026.0629.3729.87
ConsistencyLLM=QWEN-3 8B2026.0630.2530.54
FSDLLM=QWEN-3 8B2026.0630.5830.74
EntropyLLM=QWEN-3 8B2026.0630.6530.89
EntropyLLM=LLAMA-3.1 8B2026.0632.1432.44