General Reasoning on TruthfulQA
55.6AccuracyFull-step decoding
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| Full-step decodingModel=Dream-7B-Instruct, Decoding Strategy=Full-step, Evaluation Setting=Zero-shot2025.08 | 55.6 | — | — | |
| ProphetModel=Dream-7B-Instruct, Decoding Strategy=Prophet, Evaluation Setting=Zero-shot2025.08 | 53.2 | -2.4 | 1.83 | |
| ProphetModel=LLaDA-8B-Instruct, Decoding Strategy=Prophet, Evaluation Setting=Zero-shot2025.08 | 46.1 | 11.7 | 2.31 | |
| Full-step decodingModel=LLaDA-8B-Instruct, Decoding Strategy=Full-step, Evaluation Setting=Zero-shot2025.08 | 34.4 | — | — |