Selective Prediction on CommonsenseQA
0.9999PowerUCB-CLP
Evaluation Results
| Method | Links | |
|---|---|---|
| UCB-CLPLLM=OpenChat-3.5, alpha=0.22025.12 | 0.9999 | |
| UCB-CLPLLM=OpenChat-3.5, alpha=0.252025.12 | 0.9999 | |
| UCB-CLPLLM=OpenChat-3.5, alpha=0.32025.12 | 0.9999 | |
| UCB-CLPLLM=OpenChat-3.5, alpha=0.352025.12 | 0.9999 | |
| UCB-CLPLLM=OpenChat-3.5, alpha=0.42025.12 | 0.9999 | |
| UCB-CLPLLM=OpenChat-3.5, alpha=0.452025.12 | 0.9999 | |
| UCB-HFDLLM=OpenChat-3.5, alpha=0.252025.12 | 0.9999 | |
| UCB-HFDLLM=OpenChat-3.5, alpha=0.32025.12 | 0.9999 | |
| UCB-HFDLLM=OpenChat-3.5, alpha=0.352025.12 | 0.9999 | |
| UCB-HFDLLM=OpenChat-3.5, alpha=0.42025.12 | 0.9999 | |
| UCB-HFDLLM=OpenChat-3.5, alpha=0.452025.12 | 0.9999 | |
| LECLLM=OpenChat-3.5, alpha=0.22025.12 | 0.9999 | |
| LECLLM=OpenChat-3.5, alpha=0.252025.12 | 0.9999 | |
| LECLLM=OpenChat-3.5, alpha=0.32025.12 | 0.9999 | |
| LECLLM=OpenChat-3.5, alpha=0.352025.12 | 0.9999 | |
| LECLLM=OpenChat-3.5, alpha=0.42025.12 | 0.9999 | |
| LECLLM=OpenChat-3.5, alpha=0.452025.12 | 0.9999 | |
| UCB-CLPLLM=Qwen2.5-3B, alpha=0.32025.12 | 0.9999 | |
| UCB-CLPLLM=Qwen2.5-3B, alpha=0.352025.12 | 0.9999 | |
| UCB-CLPLLM=Qwen2.5-3B, alpha=0.42025.12 | 0.9999 | |
| UCB-CLPLLM=Qwen2.5-3B, alpha=0.452025.12 | 0.9999 | |
| UCB-HFDLLM=Qwen2.5-3B, alpha=0.32025.12 | 0.9999 | |
| UCB-HFDLLM=Qwen2.5-3B, alpha=0.352025.12 | 0.9999 | |
| UCB-HFDLLM=Qwen2.5-3B, alpha=0.42025.12 | 0.9999 | |
| UCB-HFDLLM=Qwen2.5-3B, alpha=0.452025.12 | 0.9999 | |
| LECLLM=Qwen2.5-3B, alpha=0.252025.12 | 0.9999 | |
| LECLLM=Qwen2.5-3B, alpha=0.32025.12 | 0.9999 | |
| LECLLM=Qwen2.5-3B, alpha=0.352025.12 | 0.9999 | |
| LECLLM=Qwen2.5-3B, alpha=0.42025.12 | 0.9999 | |
| LECLLM=Qwen2.5-3B, alpha=0.452025.12 | 0.9999 | |
| UCB-CLPLLM=Qwen2.5-7B, alpha=0.252025.12 | 0.9999 | |
| UCB-CLPLLM=Qwen2.5-7B, alpha=0.32025.12 | 0.9999 | |
| UCB-CLPLLM=Qwen2.5-7B, alpha=0.352025.12 | 0.9999 | |
| UCB-CLPLLM=Qwen2.5-7B, alpha=0.42025.12 | 0.9999 | |
| UCB-CLPLLM=Qwen2.5-7B, alpha=0.452025.12 | 0.9999 | |
| UCB-HFDLLM=Qwen2.5-7B, alpha=0.252025.12 | 0.9999 | |
| UCB-HFDLLM=Qwen2.5-7B, alpha=0.32025.12 | 0.9999 | |
| UCB-HFDLLM=Qwen2.5-7B, alpha=0.352025.12 | 0.9999 | |
| UCB-HFDLLM=Qwen2.5-7B, alpha=0.42025.12 | 0.9999 | |
| UCB-HFDLLM=Qwen2.5-7B, alpha=0.452025.12 | 0.9999 | |
| LECLLM=Qwen2.5-7B, alpha=0.252025.12 | 0.9999 | |
| LECLLM=Qwen2.5-7B, alpha=0.32025.12 | 0.9999 | |
| LECLLM=Qwen2.5-7B, alpha=0.352025.12 | 0.9999 | |
| LECLLM=Qwen2.5-7B, alpha=0.42025.12 | 0.9999 | |
| LECLLM=Qwen2.5-7B, alpha=0.452025.12 | 0.9999 | |
| UCB-CLPLLM=Qwen2.5-14B, alpha=0.22025.12 | 0.9999 | |
| UCB-CLPLLM=Qwen2.5-14B, alpha=0.252025.12 | 0.9999 | |
| UCB-CLPLLM=Qwen2.5-14B, alpha=0.32025.12 | 0.9999 | |
| UCB-CLPLLM=Qwen2.5-14B, alpha=0.352025.12 | 0.9999 | |
| UCB-CLPLLM=Qwen2.5-14B, alpha=0.42025.12 | 0.9999 | |
| UCB-CLPLLM=Qwen2.5-14B, alpha=0.452025.12 | 0.9999 | |
| UCB-HFDLLM=Qwen2.5-14B, alpha=0.22025.12 | 0.9999 | |
| UCB-HFDLLM=Qwen2.5-14B, alpha=0.252025.12 | 0.9999 | |
| UCB-HFDLLM=Qwen2.5-14B, alpha=0.32025.12 | 0.9999 | |
| UCB-HFDLLM=Qwen2.5-14B, alpha=0.352025.12 | 0.9999 | |
| UCB-HFDLLM=Qwen2.5-14B, alpha=0.42025.12 | 0.9999 | |
| UCB-HFDLLM=Qwen2.5-14B, alpha=0.452025.12 | 0.9999 | |
| LECLLM=Qwen2.5-14B, alpha=0.22025.12 | 0.9999 | |
| LECLLM=Qwen2.5-14B, alpha=0.252025.12 | 0.9999 | |
| LECLLM=Qwen2.5-14B, alpha=0.32025.12 | 0.9999 | |
| LECLLM=Qwen2.5-14B, alpha=0.352025.12 | 0.9999 | |
| LECLLM=Qwen2.5-14B, alpha=0.42025.12 | 0.9999 | |
| LECLLM=Qwen2.5-14B, alpha=0.452025.12 | 0.9999 | |
| UCB-CLPLLM=Vicuna-7B-V1.5, alpha=0.452025.12 | 0.9999 | |
| UCB-HFDLLM=Vicuna-7B-V1.5, alpha=0.452025.12 | 0.9999 | |
| LECLLM=Vicuna-7B-V1.5, alpha=0.452025.12 | 0.9999 | |
| UCB-CLPLLM=Vicuna-13B-V1.5, alpha=0.42025.12 | 0.9999 | |
| UCB-CLPLLM=Vicuna-13B-V1.5, alpha=0.452025.12 | 0.9999 | |
| UCB-HFDLLM=Vicuna-13B-V1.5, alpha=0.42025.12 | 0.9999 | |
| UCB-HFDLLM=Vicuna-13B-V1.5, alpha=0.452025.12 | 0.9999 | |
| LECLLM=Vicuna-13B-V1.5, alpha=0.42025.12 | 0.9999 | |
| LECLLM=Vicuna-13B-V1.5, alpha=0.452025.12 | 0.9999 | |
| UCB-CLPLLM=LLaMA-3.1-8B, alpha=0.32025.12 | 0.9999 | |
| UCB-CLPLLM=LLaMA-3.1-8B, alpha=0.352025.12 | 0.9999 | |
| UCB-CLPLLM=LLaMA-3.1-8B, alpha=0.42025.12 | 0.9999 | |
| UCB-CLPLLM=LLaMA-3.1-8B, alpha=0.452025.12 | 0.9999 | |
| UCB-HFDLLM=LLaMA-3.1-8B, alpha=0.32025.12 | 0.9999 | |
| UCB-HFDLLM=LLaMA-3.1-8B, alpha=0.352025.12 | 0.9999 | |
| UCB-HFDLLM=LLaMA-3.1-8B, alpha=0.42025.12 | 0.9999 | |
| UCB-HFDLLM=LLaMA-3.1-8B, alpha=0.452025.12 | 0.9999 | |
| LECLLM=LLaMA-3.1-8B, alpha=0.32025.12 | 0.9999 | |
| LECLLM=LLaMA-3.1-8B, alpha=0.352025.12 | 0.9999 | |
| LECLLM=LLaMA-3.1-8B, alpha=0.42025.12 | 0.9999 | |
| LECLLM=LLaMA-3.1-8B, alpha=0.452025.12 | 0.9999 | |
| UCB-CLPLLM=LLaMA-3.1-70B, alpha=0.252025.12 | 0.9999 | |
| UCB-CLPLLM=LLaMA-3.1-70B, alpha=0.32025.12 | 0.9999 | |
| UCB-CLPLLM=LLaMA-3.1-70B, alpha=0.352025.12 | 0.9999 | |
| UCB-CLPLLM=LLaMA-3.1-70B, alpha=0.42025.12 | 0.9999 | |
| UCB-CLPLLM=LLaMA-3.1-70B, alpha=0.452025.12 | 0.9999 | |
| UCB-HFDLLM=LLaMA-3.1-70B, alpha=0.252025.12 | 0.9999 | |
| UCB-HFDLLM=LLaMA-3.1-70B, alpha=0.32025.12 | 0.9999 | |
| UCB-HFDLLM=LLaMA-3.1-70B, alpha=0.352025.12 | 0.9999 | |
| UCB-HFDLLM=LLaMA-3.1-70B, alpha=0.42025.12 | 0.9999 | |
| UCB-HFDLLM=LLaMA-3.1-70B, alpha=0.452025.12 | 0.9999 | |
| LECLLM=LLaMA-3.1-70B, alpha=0.252025.12 | 0.9999 | |
| LECLLM=LLaMA-3.1-70B, alpha=0.32025.12 | 0.9999 | |
| LECLLM=LLaMA-3.1-70B, alpha=0.352025.12 | 0.9999 | |
| LECLLM=LLaMA-3.1-70B, alpha=0.42025.12 | 0.9999 | |
| LECLLM=LLaMA-3.1-70B, alpha=0.452025.12 | 0.9999 | |
| LECLLM=LLaMA-3.1-70B, alpha=0.22025.12 | 0.999 |