Hallucination on TruthfulQA
75.76ScoreICL
Evaluation Results
| Method | Links | |
|---|---|---|
| ICLBackbone=Qwen-2.5-7B-Instruct2026.02 | 75.76 | |
| Few-ShotBackbone=Qwen-2.5-7B-Instruct, n=22026.02 | 72.22 | |
| STEER2ADAPTBackbone=Qwen-2.5-7B-Instruct2026.02 | 71.71 | |
| REPBackbone=Qwen-2.5-7B-Instruct2026.02 | 71.06 | |
| Direct InferenceBackbone=Qwen-2.5-7B-Instruct2026.02 | 70.84 | |
| Few-ShotBackbone=Qwen-2.5-7B-Instruct, n=12026.02 | 70.63 | |
| STEER2ADAPTBackbone=Llama-3.1-8B-Instruct2026.02 | 70.54 | |
| ICLBackbone=Llama-3.1-8B-Instruct2026.02 | 70.44 | |
| REPBackbone=Llama-3.1-8B-Instruct2026.02 | 67.33 | |
| Few-ShotBackbone=Llama-3.1-8B-Instruct, n=12026.02 | 65.5 | |
| Direct InferenceBackbone=Llama-3.1-8B-Instruct2026.02 | 64.58 | |
| CAABackbone=Qwen-2.5-7B-Instruct2026.02 | 64.13 | |
| Few-ShotBackbone=Llama-3.1-8B-Instruct, n=22026.02 | 62.16 | |
| CAABackbone=Llama-3.1-8B-Instruct2026.02 | 59.62 | |
| REPBackbone=Mistral-7B-Instruct-v0.12026.02 | 56.45 | |
| ICLBackbone=Mistral-7B-Instruct-v0.12026.02 | 54.94 | |
| STEER2ADAPTBackbone=Mistral-7B-Instruct-v0.12026.02 | 54.02 | |
| SPINIteration=Iter32025.12 | 52.85 | |
| SPINIteration=Iter12025.12 | 52.12 | |
| SPINIteration=Iter22025.12 | 51.77 | |
| SPINIteration=Iter42025.12 | 51.73 | |
| CAABackbone=Mistral-7B-Instruct-v0.12026.02 | 50.8 | |
| SPINIteration=Iter02025.12 | 48.93 | |
| SPACEIteration=Iter12025.12 | 48.74 | |
| SPACEIteration=Iter32025.12 | 48.51 | |
| SPACEIteration=Iter42025.12 | 48.38 | |
| SPACEIteration=Iter22025.12 | 48.32 | |
| S-IPOIteration=Iter12025.12 | 47.64 | |
| S-IPOIteration=Iter02025.12 | 46.93 | |
| Direct InferenceBackbone=Mistral-7B-Instruct-v0.12026.02 | 46.18 | |
| SPACEIteration=Iter02025.12 | 46.05 | |
| S-IPOIteration=Iter42025.12 | 46 | |
| S-IPOIteration=Iter32025.12 | 45.95 | |
| S-IPOIteration=Iter22025.12 | 45.92 | |
| S-SimPOIteration=Iter12025.12 | 43.93 | |
| Zephyr-7BIteration=N/A2025.12 | 43.67 | |
| S-SimPOIteration=Iter22025.12 | 43.08 | |
| S-SimPOIteration=Iter32025.12 | 42.59 | |
| S-SimPOIteration=Iter42025.12 | 42.56 | |
| S-SimPOIteration=Iter02025.12 | 41.46 | |
| Few-ShotBackbone=Mistral-7B-Instruct-v0.1, n=12026.02 | 35.38 | |
| Few-ShotBackbone=Mistral-7B-Instruct-v0.1, n=22026.02 | 34.33 |