oumi-ai/oumi-groundedness-benchmark
收藏Hugging Face2025-04-04 更新2025-04-12 收录
下载链接：
https://hf-mirror.com/datasets/oumi-ai/oumi-groundedness-benchmark
下载链接
链接失效反馈官方服务：
资源简介：
oumi-groundedness-benchmark是一个文本数据集，旨在评估语言模型在声明验证/虚构检测方面的性能。该数据集通过在Llama-3.1-405B-Instruct模型上运行Oumi推理生成提示和响应。oumi-groundedness-benchmark用于正确评估HallOumi-8B模型，该模型在Macro F1指标上达到了77.2%，超过了Claude Sonnet 3.5、OpenAI o1等最新模型。
oumi-groundedness-benchmark is a text dataset designed to evaluate language models for Claim Verification / Hallucination Detection. Prompts and responses were produced synthetically from Llama-3.1-405B-Instruct. oumi-groundedness-benchmark was used to properly evaluate HallOumi-8B, which achieves 77.2% Macro F1, outperforming SOTA models such as Claude Sonnet 3.5, OpenAI o1, etc.
提供机构：
oumi-ai
