Text Classification on Yahoo! Answers (ECE, BS)
3.04ECECaliDist_verbalized (As.)
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| CaliDist_verbalized (As.)LLM=QWEN-2 7B2026.06 | 3.04 | 19.82 | |
| CaliDist_verbalized (Pr.)LLM=QWEN-2 7B2026.06 | 3.29 | 19.71 | |
| CaliDist_verbalized (Pr.)LLM=LLAMA-3.1 8B2026.06 | 3.51 | 18.47 | |
| CaliDist (As.)LLM=QWEN-3 8B2026.06 | 3.73 | 18.47 | |
| CaliDist_verbalized (Sa.)LLM=QWEN-2 7B2026.06 | 3.85 | 19.97 | |
| CaliDist (As.)LLM=LLAMA-3.1 8B2026.06 | 4.11 | 16.88 | |
| CaliDist (Pr.)LLM=LLAMA-3.1 8B2026.06 | 4.56 | 16.89 | |
| CaliDist_verbalized (As.)LLM=LLAMA-3.1 8B2026.06 | 4.96 | 18.65 | |
| CaliDist_verbalized (Sa.)LLM=LLAMA-3.1 8B2026.06 | 5.07 | 18.75 | |
| CaliDist (Pr.)LLM=QWEN-3 8B2026.06 | 5.11 | 18.49 | |
| CaliDist (Sa.)LLM=LLAMA-3.1 8B2026.06 | 5.85 | 17.2 | |
| CaliDist (Sa.)LLM=QWEN-3 8B2026.06 | 6.53 | 19.62 | |
| Temperature ScalingLLM=LLAMA-3.1 8B2026.06 | 12.64 | 17.72 | |
| VerbalizedLLM=QWEN-2 7B2026.06 | 13.1 | 21.33 | |
| VerbalizedLLM=LLAMA-3.1 8B2026.06 | 16.82 | 21.1 | |
| VanillaLLM=LLAMA-3.1 8B2026.06 | 19.74 | 21.18 | |
| EntropyLLM=LLAMA-3.1 8B2026.06 | 19.89 | 21.25 | |
| FSDLLM=LLAMA-3.1 8B2026.06 | 21.04 | 22.65 | |
| ConsistencyLLM=LLAMA-3.1 8B2026.06 | 25.01 | 24.82 | |
| Temperature ScalingLLM=QWEN-3 8B2026.06 | 28.45 | 28.36 | |
| EntropyLLM=QWEN-3 8B2026.06 | 29.05 | 29.21 | |
| VanillaLLM=QWEN-3 8B2026.06 | 29.42 | 29.38 | |
| FSDLLM=QWEN-3 8B2026.06 | 29.52 | 29.57 | |
| ConsistencyLLM=QWEN-3 8B2026.06 | 29.86 | 29.89 |