Paraphrase Identification on PPDB
2.28ECECaliDist_verbalized (Sa.)
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| CaliDist_verbalized (Sa.)LLM=LLAMA-3.1 8B2026.06 | 2.28 | 17.52 | |
| CaliDist (Sa.)LLM=QWEN-3 8B2026.06 | 2.39 | 21.38 | |
| CaliDist_verbalized (Sa.)LLM=QWEN-2 7B2026.06 | 2.61 | 20.84 | |
| CaliDist (Pr.)LLM=LLAMA-3.1 8B2026.06 | 3.75 | 18.91 | |
| CaliDist (Pr.)LLM=QWEN-3 8B2026.06 | 4.57 | 24.72 | |
| CaliDist_verbalized (As.)LLM=LLAMA-3.1 8B2026.06 | 5.07 | 19.55 | |
| Temperature ScalingLLM=LLAMA-3.1 8B2026.06 | 5.12 | 18.98 | |
| CaliDist (Sa.)LLM=LLAMA-3.1 8B2026.06 | 5.42 | 18.72 | |
| CaliDist (As.)LLM=QWEN-3 8B2026.06 | 5.87 | 21.93 | |
| CaliDist_verbalized (As.)LLM=QWEN-2 7B2026.06 | 6.86 | 21.51 | |
| CaliDist (As.)LLM=LLAMA-3.1 8B2026.06 | 7.02 | 19.16 | |
| CaliDist_verbalized (Pr.)LLM=LLAMA-3.1 8B2026.06 | 7.7 | 19.37 | |
| CaliDist_verbalized (Pr.)LLM=QWEN-2 7B2026.06 | 9.34 | 23.86 | |
| CaliDist_verbalized (Sa.)LLM=GEMMA-3 4B2026.06 | 12.29 | 22.89 | |
| VerbalizedLLM=GEMMA-3 4B2026.06 | 12.77 | 22.58 | |
| CaliDist_verbalized (As.)LLM=GEMMA-3 4B2026.06 | 12.98 | 22.79 | |
| VanillaLLM=LLAMA-3.1 8B2026.06 | 13.5 | 20.87 | |
| CaliDist_verbalized (Pr.)LLM=GEMMA-3 4B2026.06 | 13.52 | 22.7 | |
| CaliDist_verbalized (Sa.)LLM=PHI-4-MINI 3.8B2026.06 | 14.44 | 24.48 | |
| VerbalizedLLM=LLAMA-3.1 8B2026.06 | 14.63 | 19.83 | |
| CaliDist_verbalized (Pr.)LLM=PHI-4-MINI 3.8B2026.06 | 15.15 | 24.43 | |
| CaliDist_verbalized (As.)LLM=PHI-4-MINI 3.8B2026.06 | 15.81 | 24.63 | |
| VerbalizedLLM=PHI-4-MINI 3.8B2026.06 | 15.86 | 24.5 | |
| ConsistencyLLM=LLAMA-3.1 8B2026.06 | 19.35 | 23.34 | |
| FSDLLM=LLAMA-3.1 8B2026.06 | 23.26 | 24.78 | |
| EntropyLLM=LLAMA-3.1 8B2026.06 | 26.68 | 27.88 | |
| VerbalizedLLM=QWEN-2 7B2026.06 | 31.64 | 32.32 | |
| Temperature ScalingLLM=QWEN-3 8B2026.06 | 38.8 | 39.05 | |
| EntropyLLM=QWEN-3 8B2026.06 | 38.96 | 38.87 | |
| FSDLLM=QWEN-3 8B2026.06 | 39.1 | 39.14 | |
| VanillaLLM=QWEN-3 8B2026.06 | 39.46 | 39.66 | |
| ConsistencyLLM=QWEN-3 8B2026.06 | 39.47 | 39.46 |