Question Answering on WQ (PRR)
62.8PRROurs
Evaluation results
| Method | Links | |
|---|---|---|
| OursLLM Model=Qwen2.5-7B2026.03 | 62.8 | |
| CCPLLM Model=Qwen2.5-7B2026.03 | 60.2 | |
| OursLLM Model=Qwen2.5-14B2026.03 | 58.7 | |
| OursLLM Model=Qwen2.5-3B2026.03 | 57.3 | |
| OursLLM Model=Llama-3-8B2026.03 | 57 | |
| sentenceSARLLM Model=Qwen2.5-7B2026.03 | 56.8 | |
| OursLLM Model=Llama-2-7B2026.03 | 56.8 | |
| SARLLM Model=Qwen2.5-7B2026.03 | 56.2 | |
| Predictive EntropyLLM Model=Qwen2.5-7B2026.03 | 56 | |
| CCPLLM Model=Qwen2.5-14B2026.03 | 55.3 | |
| sentenceSARLLM Model=Qwen2.5-14B2026.03 | 54.8 | |
| SARLLM Model=Qwen2.5-14B2026.03 | 54.6 | |
| LN EntropyLLM Model=Qwen2.5-7B2026.03 | 54.3 | |
| sentenceSARLLM Model=Llama-3-8B2026.03 | 54.3 | |
| SARLLM Model=Llama-3-8B2026.03 | 54 | |
| OursLLM Model=Mistral-7B2026.03 | 54 | |
| PerplexityLLM Model=Qwen2.5-14B2026.03 | 53.9 | |
| Predictive EntropyLLM Model=Qwen2.5-14B2026.03 | 53.9 | |
| Predictive EntropyLLM Model=Llama-3-8B2026.03 | 53.3 | |
| LN EntropyLLM Model=Qwen2.5-14B2026.03 | 53.2 | |
| LN EntropyLLM Model=Llama-3-8B2026.03 | 53 | |
| CCPLLM Model=Qwen2.5-3B2026.03 | 52.8 | |
| EigVLLM Model=Mistral-7B2026.03 | 52.3 | |
| EigVLLM Model=Qwen2.5-14B2026.03 | 52 | |
| PerplexityLLM Model=Qwen2.5-7B2026.03 | 51.9 | |
| tokenSARLLM Model=Qwen2.5-14B2026.03 | 51.5 | |
| sentenceSARLLM Model=Mistral-7B2026.03 | 51.2 | |
| CCPLLM Model=Llama-2-7B2026.03 | 51.2 | |
| CCPLLM Model=Llama-3-8B2026.03 | 51.2 | |
| sentenceSARLLM Model=Llama-2-7B2026.03 | 50.7 | |
| CCPLLM Model=Mistral-7B2026.03 | 50.5 | |
| sentenceSARLLM Model=Qwen2.5-3B2026.03 | 50.4 | |
| P(True)LLM Model=Qwen2.5-7B2026.03 | 50.3 | |
| EigVLLM Model=Qwen2.5-7B2026.03 | 50.1 | |
| Predictive EntropyLLM Model=Mistral-7B2026.03 | 50 | |
| EigVLLM Model=Llama-2-7B2026.03 | 50 | |
| Predictive EntropyLLM Model=Llama-2-7B2026.03 | 49.6 | |
| Predictive EntropyLLM Model=Qwen2.5-3B2026.03 | 49.5 | |
| DegLLM Model=Qwen2.5-7B2026.03 | 49.1 | |
| DegLLM Model=Mistral-7B2026.03 | 49 | |
| tokenSARLLM Model=Qwen2.5-7B2026.03 | 47.5 | |
| DegLLM Model=Qwen2.5-14B2026.03 | 47.5 | |
| P(True)LLM Model=Qwen2.5-14B2026.03 | 47.3 | |
| EigVLLM Model=Qwen2.5-3B2026.03 | 47 | |
| SARLLM Model=Qwen2.5-3B2026.03 | 46.9 | |
| SARLLM Model=Mistral-7B2026.03 | 46.6 | |
| LN EntropyLLM Model=Llama-2-7B2026.03 | 46.3 | |
| SARLLM Model=Llama-2-7B2026.03 | 46.3 | |
| EigVLLM Model=Llama-3-8B2026.03 | 46.2 | |
| DegLLM Model=Llama-2-7B2026.03 | 46.2 | |
| Semantic EntropyLLM Model=Qwen2.5-7B2026.03 | 45.5 | |
| DegLLM Model=Llama-3-8B2026.03 | 45.5 | |
| PerplexityLLM Model=Llama-3-8B2026.03 | 45.2 | |
| Eigen ScoreLLM Model=Llama-3-8B2026.03 | 45 | |
| EccLLM Model=Mistral-7B2026.03 | 44.8 | |
| Eigen ScoreLLM Model=Mistral-7B2026.03 | 44.5 | |
| LN EntropyLLM Model=Mistral-7B2026.03 | 44.3 | |
| LN EntropyLLM Model=Qwen2.5-3B2026.03 | 44.2 | |
| Semantic EntropyLLM Model=Qwen2.5-3B2026.03 | 43.9 | |
| ConULLM Model=Qwen2.5-7B2026.03 | 43.4 | |
| Eigen ScoreLLM Model=Qwen2.5-7B2026.03 | 43.4 | |
| Semantic EntropyLLM Model=Qwen2.5-14B2026.03 | 43.3 | |
| EccLLM Model=Qwen2.5-14B2026.03 | 43.1 | |
| ConULLM Model=Mistral-7B2026.03 | 42.8 | |
| Eigen ScoreLLM Model=Qwen2.5-14B2026.03 | 42.6 | |
| DegLLM Model=Qwen2.5-3B2026.03 | 42.5 | |
| ConULLM Model=Qwen2.5-14B2026.03 | 42.4 | |
| EccLLM Model=Qwen2.5-7B2026.03 | 41.9 | |
| Semantic EntropyLLM Model=Mistral-7B2026.03 | 41.8 | |
| EccLLM Model=Llama-2-7B2026.03 | 41.6 | |
| tokenSARLLM Model=Llama-3-8B2026.03 | 41.5 | |
| Eigen ScoreLLM Model=Llama-2-7B2026.03 | 41.4 | |
| ConULLM Model=Qwen2.5-3B2026.03 | 40.1 | |
| PerplexityLLM Model=Llama-2-7B2026.03 | 39.9 | |
| Semantic EntropyLLM Model=Llama-2-7B2026.03 | 39.9 | |
| EccLLM Model=Llama-3-8B2026.03 | 39.6 | |
| EccLLM Model=Qwen2.5-3B2026.03 | 39.3 | |
| PerplexityLLM Model=Mistral-7B2026.03 | 38.8 | |
| Semantic EntropyLLM Model=Llama-3-8B2026.03 | 38.1 | |
| Eigen ScoreLLM Model=Qwen2.5-3B2026.03 | 36.9 | |
| ConULLM Model=Llama-2-7B2026.03 | 36.2 | |
| PerplexityLLM Model=Qwen2.5-3B2026.03 | 33.1 | |
| tokenSARLLM Model=Llama-2-7B2026.03 | 32.8 | |
| tokenSARLLM Model=Qwen2.5-3B2026.03 | 31 | |
| tokenSARLLM Model=Mistral-7B2026.03 | 29.9 | |
| ConULLM Model=Llama-3-8B2026.03 | 29.2 | |
| P(True)LLM Model=Qwen2.5-3B2026.03 | 28 | |
| P(True)LLM Model=Llama-3-8B2026.03 | 17 | |
| P(True)LLM Model=Mistral-7B2026.03 | 16.8 | |
| P(True)LLM Model=Llama-2-7B2026.03 | 9 |