TruthfulQA
更新于 2025-07-22
1,023 次浏览
问题数量
817
发布机构
Google
评测类别
真实性评估
评测指标
Accuracy
支持语言
英语
难度等级
Advanced
简介
一个包含 817 个问题的基准，旨在评估模型是否能够生成真实且准确的答案，而不是编造信息。
相关资源
TruthfulQA Model Score Leaderboard
Source: DataLearnerAI
Data sourced primarily from official releases (GitHub, Hugging Face, papers), then benchmark leaderboards, then third-party evaluators. Learn about our data methodology
模式说明:
normal
thinking
low
medium
high
deeper thinking
parallel_thinking
图表加载中...
Latest TruthfulQA model rankings and full benchmark leaderboard
Browse the latest scores, model modes, release dates, and parameter sizes for TruthfulQA.
Model release cutoff:
TruthfulQA详细排名数据表格
| 排名 | 模型 | |||
|---|---|---|---|---|
| 1 | Qwen2.5-72B Standard Mode | 60.40 | 2024-09-18 | 727 |
| 2 | Gemini 1.5 Pro Standard Mode | 0.00 | 2024-02-15 | 未知 |
| 3 | Llama3.1-405B Instruct Standard Mode | 0.00 | 2024-07-23 | 4050 |
| 4 | Amazon Nova Pro Standard Mode | 0.00 | 2024-12-03 | 未知 |