🏅 OmniEval Leaderboard
Please contact us if you would like to submit your model to this leaderboard. Email: wangshuting@ruc.edu.cn
如果您想将您的模型提交到此排行榜,请联系我们。邮箱:wangshuting@ruc.edu.cn
{
- "headers": [
- "Type Symbol",
- "Retrieval Model Plain",
- "Generative Model Plain",
- "Retrieval Model",
- "Generative Model",
- "Gen Average ⬆️",
- "Ret Average ⬆️",
- "MRR ⬆️",
- "MAP ⬆️",
- "EM ⬆️",
- "F1 ⬆️",
- "Rouge-1 ⬆️",
- "Rouge-2 ⬆️",
- "Rouge-L ⬆️",
- "ACC ⬆️",
- "COMP ⬆️",
- "HAL ⬇️",
- "UTIL ⬆️",
- "MACC ⬆️",
- "Type",
- "Gen#Params (B)",
- "Ret#Params (B)"
- "data": [
- [
- "🟢",
- "Alibaba-NLP/gte-Qwen2-1.5B-instruct",
- "Qwen/Qwen2.5-72B-Instruct",
- "<a target="_blank" href="https://huggingface.co/Alibaba-NLP/gte-Qwen2-1.5B-instruct" style="color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;">Alibaba-NLP/gte-Qwen2-1.5B-instruct</a>",
- "<a target="_blank" href="https://huggingface.co/Qwen/Qwen2.5-72B-Instruct" style="color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;">Qwen/Qwen2.5-72B-Instruct</a>",
- 35.95,
- 44.43,
- 45.74,
- 44.43,
- 0.28,
- 39.19,
- 36.14,
- 20.78,
- 25.28,
- 44.76,
- 61.9,
- 6.18,
- 55.76,
- 35.95,
- "open-source",
- 72.7,
- 1.78
- [
- "🟢",
- "BAAI/bge-large-zh",
- "Qwen/Qwen2.5-72B-Instruct",
- "<a target="_blank" href="https://huggingface.co/BAAI/bge-large-zh" style="color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;">BAAI/bge-large-zh</a>",
- "<a target="_blank" href="https://huggingface.co/Qwen/Qwen2.5-72B-Instruct" style="color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;">Qwen/Qwen2.5-72B-Instruct</a>",
- 35.45,
- 41.53,
- 42.52,
- 41.53,
- 0.34,
- 38.7,
- 35.04,
- 19.63,
- 24.35,
- 43.25,
- 62.24,
- 7.18,
- 53.67,
- 35.45,
- "open-source",
- 72.7,
- 0.33
- [
- "🟢",
- "BAAI/bge-m3",
- "Qwen/Qwen2.5-72B-Instruct",
- "<a target="_blank" href="https://huggingface.co/BAAI/bge-m3" style="color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;">BAAI/bge-m3</a>",
- "<a target="_blank" href="https://huggingface.co/Qwen/Qwen2.5-72B-Instruct" style="color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;">Qwen/Qwen2.5-72B-Instruct</a>",
- 34.72,
- 41.52,
- 42.36,
- 41.52,
- 0.4,
- 39.13,
- 35.73,
- 20.43,
- 25.17,
- 44.5,
- 62.08,
- 7.09,
- 54.1,
- 34.72,
- "open-source",
- 72.7,
- 0.5
- [
- "🟢",
- "Alibaba-NLP/gte-Qwen2-1.5B-instruct",
- "meta-llama/Meta-Llama-3.1-70B-Instruct",
- "<a target="_blank" href="https://huggingface.co/Alibaba-NLP/gte-Qwen2-1.5B-instruct" style="color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;">Alibaba-NLP/gte-Qwen2-1.5B-instruct</a>",
- "<a target="_blank" href="https://huggingface.co/meta-llama/Meta-Llama-3.1-70B-Instruct" style="color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;">meta-llama/Meta-Llama-3.1-70B-Instruct</a>",
- 32.68,
- 44.43,
- 45.74,
- 44.43,
- 5.13,
- 50.42,
- 42.53,
- 25.01,
- 33.9,
- 44.33,
- 57.45,
- 11.31,
- 47.64,
- 32.68,
- "open-source",
- 70.6,
- 1.78
- [
- "🟢",
- "Alibaba-NLP/gte-Qwen2-1.5B-instruct",
- "deepseek-ai/DeepSeek-V2-Chat-0628",
- "<a target="_blank" href="https://huggingface.co/Alibaba-NLP/gte-Qwen2-1.5B-instruct" style="color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;">Alibaba-NLP/gte-Qwen2-1.5B-instruct</a>",
- "<a target="_blank" href="https://huggingface.co/deepseek-ai/DeepSeek-V2-Chat-0628" style="color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;">deepseek-ai/DeepSeek-V2-Chat-0628</a>",
- 32.13,
- 44.43,
- 45.74,
- 44.43,
- 0.51,
- 43.53,
- 39.11,
- 23.49,
- 29.16,
- 42.34,
- 60.06,
- 7.5,
- 51.6,
- 32.13,
- "open-source",
- 236,
- 1.78
- [
- "🟢",
- "jinaai/jina-embeddings-v2-base-zh",
- "Qwen/Qwen2.5-72B-Instruct",
- "<a target="_blank" href="https://huggingface.co/jinaai/jina-embeddings-v2-base-zh" style="color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;">jinaai/jina-embeddings-v2-base-zh</a>",
- "<a target="_blank" href="https://huggingface.co/Qwen/Qwen2.5-72B-Instruct" style="color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;">Qwen/Qwen2.5-72B-Instruct</a>",
- 30.73,
- 34.58,
- 35.33,
- 34.58,
- 0.4,
- 38.21,
- 33.74,
- 18.43,
- 23.41,
- 40.89,
- 59.3,
- 8.86,
- 51.63,
- 30.73,
- "open-source",
- 72.7,
- 0.16
- [
- "🟢",
- "intfloat/e5-mistral-7b-instruct",
- "Qwen/Qwen2.5-72B-Instruct",
- "<a target="_blank" href="https://huggingface.co/intfloat/e5-mistral-7b-instruct" style="color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;">intfloat/e5-mistral-7b-instruct</a>",
- "<a target="_blank" href="https://huggingface.co/Qwen/Qwen2.5-72B-Instruct" style="color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;">Qwen/Qwen2.5-72B-Instruct</a>",
- 30.5,
- 37.69,
- 38.65,
- 37.69,
- 0.23,
- 37.87,
- 34.04,
- 18.98,
- 23.62,
- 40.69,
- 59.55,
- 7.92,
- 51.17,
- 30.5,
- "open-source",
- 72.7,
- 7.11
- [
- "🟢",
- "Alibaba-NLP/gte-Qwen2-1.5B-instruct",
- "01ai/Yi-1.5-34B-Chat-16K",
- "<a target="_blank" href="https://huggingface.co/Alibaba-NLP/gte-Qwen2-1.5B-instruct" style="color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;">Alibaba-NLP/gte-Qwen2-1.5B-instruct</a>",
- "<a target="_blank" href="https://huggingface.co/01ai/Yi-1.5-34B-Chat-16K" style="color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;">01ai/Yi-1.5-34B-Chat-16K</a>",
- 18.56,
- 44.43,
- 45.74,
- 44.43,
- 0,
- 15.83,
- 21.95,
- 9.92,
- 8.87,
- 33.66,
- 58.21,
- 6.48,
- 42.34,
- 18.56,
- "open-source",
- 34.4,
- 1.78
- [
- "🟢",
- "CLOSE",
- "Qwen/Qwen2.5-72B-Instruct",
- "CLOSE",
- "<a target="_blank" href="https://huggingface.co/Qwen/Qwen2.5-72B-Instruct" style="color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;">Qwen/Qwen2.5-72B-Instruct</a>",
- 12.41,
- 0,
- 0,
- 0,
- 0.06,
- 41.91,
- 29.9,
- 12.05,
- 20.82,
- 34.05,
- 57.54,
- 0,
- 0,
- 12.41,
- "open-source",
- 72.7,
- 0
- [
- "🟢",
- "CLOSE",
- "deepseek-ai/DeepSeek-V2-Chat-0628",
- "CLOSE",
- "<a target="_blank" href="https://huggingface.co/deepseek-ai/DeepSeek-V2-Chat-0628" style="color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;">deepseek-ai/DeepSeek-V2-Chat-0628</a>",
- 11.53,
- 0,
- 0,
- 0,
- 0.23,
- 43.53,
- 31.51,
- 12.93,
- 22.5,
- 33.06,
- 55.41,
- 0,
- 0,
- 11.53,
- "open-source",
- 236,
- 0
- [
- "🟢",
- "CLOSE",
- "01ai/Yi-1.5-34B-Chat-16K",
- "CLOSE",
- "<a target="_blank" href="https://huggingface.co/01ai/Yi-1.5-34B-Chat-16K" style="color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;">01ai/Yi-1.5-34B-Chat-16K</a>",
- 7.49,
- 0,
- 0,
- 0,
- 0,
- 11.61,
- 15.61,
- 4.59,
- 4.97,
- 14.61,
- 49.87,
- 0,
- 0,
- 7.49,
- "open-source",
- 34.4,
- 0
- [
- "🟢",
- "CLOSE",
- "meta-llama/Meta-Llama-3.1-70B-Instruct",
- "CLOSE",
- "<a target="_blank" href="https://huggingface.co/meta-llama/Meta-Llama-3.1-70B-Instruct" style="color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;">meta-llama/Meta-Llama-3.1-70B-Instruct</a>",
- 6.59,
- 0,
- 0,
- 0,
- 0.06,
- 41.83,
- 30.61,
- 12.05,
- 21.95,
- 28.59,
- 51.33,
- 0,
- 0,
- 6.59,
- "open-source",
- 70.6,
- 0
- [
- "metadata": null
{
- "headers": [
- "Type Symbol",
- "Retrieval Model Plain",
- "Generative Model Plain",
- "Retrieval Model",
- "Generative Model",
- "Gen Average ⬆️",
- "Ret Average ⬆️",
- "MRR ⬆️",
- "MAP ⬆️",
- "EM ⬆️",
- "F1 ⬆️",
- "Rouge-1 ⬆️",
- "Rouge-2 ⬆️",
- "Rouge-L ⬆️",
- "ACC ⬆️",
- "COMP ⬆️",
- "HAL ⬇️",
- "UTIL ⬆️",
- "MACC ⬆️",
- "Type",
- "Gen#Params (B)",
- "Ret#Params (B)"
- "data": [
- [
- "🟢",
- "Alibaba-NLP/gte-Qwen2-1.5B-instruct",
- "Qwen/Qwen2.5-72B-Instruct",
- "<a target="_blank" href="https://huggingface.co/Alibaba-NLP/gte-Qwen2-1.5B-instruct" style="color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;">Alibaba-NLP/gte-Qwen2-1.5B-instruct</a>",
- "<a target="_blank" href="https://huggingface.co/Qwen/Qwen2.5-72B-Instruct" style="color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;">Qwen/Qwen2.5-72B-Instruct</a>",
- 32.93,
- 43.7,
- 44.91,
- 43.7,
- 0.43,
- 25.63,
- 26.86,
- 14.44,
- 17.78,
- 43.26,
- 62.56,
- 4.67,
- 56.13,
- 32.93,
- "open-source",
- 72.7,
- 1.78
- [
- "🟢",
- "Alibaba-NLP/gte-Qwen2-1.5B-instruct",
- "deepseek-ai/DeepSeek-V2-Chat-0628",
- "<a target="_blank" href="https://huggingface.co/Alibaba-NLP/gte-Qwen2-1.5B-instruct" style="color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;">Alibaba-NLP/gte-Qwen2-1.5B-instruct</a>",
- "<a target="_blank" href="https://huggingface.co/deepseek-ai/DeepSeek-V2-Chat-0628" style="color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;">deepseek-ai/DeepSeek-V2-Chat-0628</a>",
- 31.75,
- 43.7,
- 44.91,
- 43.7,
- 0.47,
- 33,
- 31.14,
- 17.52,
- 22.79,
- 40.99,
- 60.72,
- 6.34,
- 51.97,
- 31.75,
- "open-source",
- 236,
- 1.78
- [
- "🟢",
- "BAAI/bge-large-zh",
- "Qwen/Qwen2.5-72B-Instruct",
- "<a target="_blank" href="https://huggingface.co/BAAI/bge-large-zh" style="color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;">BAAI/bge-large-zh</a>",
- "<a target="_blank" href="https://huggingface.co/Qwen/Qwen2.5-72B-Instruct" style="color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;">Qwen/Qwen2.5-72B-Instruct</a>",
- 31.24,
- 37.77,
- 38.65,
- 37.77,
- 0.32,
- 25.41,
- 25.83,
- 13.27,
- 16.93,
- 40.8,
- 60.48,
- 5.97,
- 51.94,
- 31.24,
- "open-source",
- 72.7,
- 0.33
- [
- "🟢",
- "Alibaba-NLP/gte-Qwen2-1.5B-instruct",
- "meta-llama/Meta-Llama-3.1-70B-Instruct",
- "<a target="_blank" href="https://huggingface.co/Alibaba-NLP/gte-Qwen2-1.5B-instruct" style="color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;">Alibaba-NLP/gte-Qwen2-1.5B-instruct</a>",
- "<a target="_blank" href="https://huggingface.co/meta-llama/Meta-Llama-3.1-70B-Instruct" style="color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;">meta-llama/Meta-Llama-3.1-70B-Instruct</a>",
- 30.88,
- 43.7,
- 44.91,
- 43.7,
- 4.03,
- 48.1,
- 39.95,
- 23.05,
- 32.35,
- 43.98,
- 59.26,
- 7.92,
- 47.54,
- 30.88,
- "open-source",
- 70.6,
- 1.78
- [
- "🟢",
- "BAAI/bge-m3",
- "Qwen/Qwen2.5-72B-Instruct",
- "<a target="_blank" href="https://huggingface.co/BAAI/bge-m3" style="color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;">BAAI/bge-m3</a>",
- "<a target="_blank" href="https://huggingface.co/Qwen/Qwen2.5-72B-Instruct" style="color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;">Qwen/Qwen2.5-72B-Instruct</a>",
- 30.6,
- 39.61,
- 40.57,
- 39.61,
- 0.32,
- 25.93,
- 26.36,
- 13.85,
- 17.46,
- 40.91,
- 60.92,
- 6.34,
- 52.03,
- 30.6,
- "open-source",
- 72.7,
- 0.5
- [
- "🟢",
- "intfloat/e5-mistral-7b-instruct",
- "Qwen/Qwen2.5-72B-Instruct",
- "<a target="_blank" href="https://huggingface.co/intfloat/e5-mistral-7b-instruct" style="color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;">intfloat/e5-mistral-7b-instruct</a>",
- "<a target="_blank" href="https://huggingface.co/Qwen/Qwen2.5-72B-Instruct" style="color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;">Qwen/Qwen2.5-72B-Instruct</a>",
- 28.78,
- 34.17,
- 35.08,
- 34.17,
- 0.33,
- 24.9,
- 25.21,
- 12.71,
- 16.36,
- 39.74,
- 59.33,
- 6.48,
- 51.51,
- 28.78,
- "open-source",
- 72.7,
- 7.11
- [
- "🟢",
- "jinaai/jina-embeddings-v2-base-zh",
- "Qwen/Qwen2.5-72B-Instruct",
- "<a target="_blank" href="https://huggingface.co/jinaai/jina-embeddings-v2-base-zh" style="color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;">jinaai/jina-embeddings-v2-base-zh</a>",
- "<a target="_blank" href="https://huggingface.co/Qwen/Qwen2.5-72B-Instruct" style="color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;">Qwen/Qwen2.5-72B-Instruct</a>",
- 28.37,
- 33.95,
- 34.69,
- 33.95,
- 0.4,
- 25.53,
- 25.32,
- 12.8,
- 16.62,
- 39.08,
- 59.81,
- 7.94,
- 50.78,
- 28.37,
- "open-source",
- 72.7,
- 0.16
- [
- "🟢",
- "Alibaba-NLP/gte-Qwen2-1.5B-instruct",
- "01ai/Yi-1.5-34B-Chat-16K",
- "<a target="_blank" href="https://huggingface.co/Alibaba-NLP/gte-Qwen2-1.5B-instruct" style="color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;">Alibaba-NLP/gte-Qwen2-1.5B-instruct</a>",
- "<a target="_blank" href="https://huggingface.co/01ai/Yi-1.5-34B-Chat-16K" style="color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;">01ai/Yi-1.5-34B-Chat-16K</a>",
- 16.82,
- 43.7,
- 44.91,
- 43.7,
- 0,
- 9.58,
- 16.51,
- 6.7,
- 5.93,
- 34.02,
- 57.78,
- 5.97,
- 42.29,
- 16.82,
- "open-source",
- 34.4,
- 1.78
- [
- "🟢",
- "CLOSE",
- "Qwen/Qwen2.5-72B-Instruct",
- "CLOSE",
- "<a target="_blank" href="https://huggingface.co/Qwen/Qwen2.5-72B-Instruct" style="color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;">Qwen/Qwen2.5-72B-Instruct</a>",
- 12.56,
- 0,
- 0,
- 0,
- 0.03,
- 32.22,
- 23.52,
- 8.06,
- 16.07,
- 37.88,
- 60.17,
- 0,
- 0,
- 12.56,
- "open-source",
- 72.7,
- 0
- [
- "🟢",
- "CLOSE",
- "deepseek-ai/DeepSeek-V2-Chat-0628",
- "CLOSE",
- "<a target="_blank" href="https://huggingface.co/deepseek-ai/DeepSeek-V2-Chat-0628" style="color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;">deepseek-ai/DeepSeek-V2-Chat-0628</a>",
- 11.21,
- 0,
- 0,
- 0,
- 0.12,
- 37.09,
- 25.71,
- 9.09,
- 18.61,
- 35.87,
- 57.55,
- 0,
- 0,
- 11.21,
- "open-source",
- 236,
- 0
- [
- "🟢",
- "CLOSE",
- "01ai/Yi-1.5-34B-Chat-16K",
- "CLOSE",
- "<a target="_blank" href="https://huggingface.co/01ai/Yi-1.5-34B-Chat-16K" style="color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;">01ai/Yi-1.5-34B-Chat-16K</a>",
- 6.93,
- 0,
- 0,
- 0,
- 0,
- 6.73,
- 12.78,
- 3.21,
- 3.26,
- 15.73,
- 50.63,
- 0,
- 0,
- 6.93,
- "open-source",
- 34.4,
- 0
- [
- "🟢",
- "CLOSE",
- "meta-llama/Meta-Llama-3.1-70B-Instruct",
- "CLOSE",
- "<a target="_blank" href="https://huggingface.co/meta-llama/Meta-Llama-3.1-70B-Instruct" style="color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;">meta-llama/Meta-Llama-3.1-70B-Instruct</a>",
- 6.77,
- 0,
- 0,
- 0,
- 0.09,
- 39.89,
- 26.8,
- 9.29,
- 19.93,
- 32.38,
- 52.84,
- 0,
- 0,
- 6.77,
- "open-source",
- 70.6,
- 0
- [
- "metadata": null
Leaderboard Information
We introduce an omnidirectional and automatic RAG benchmark, OmniEval: An Omnidirectional and Automatic RAG Evaluation Benchmark in Financial Domain, in the financial domain. Our benchmark is characterized by its multi-dimensional evaluation framework, including:
- a matrix-based RAG scenario evaluation system that categorizes queries into five task classes and 16 financial topics, leading to a structured assessment of diverse query scenarios;
- a multi-dimensional evaluation data generation approach, which combines GPT-4-based automatic generation and human annotation, achieving an 87.47% acceptance ratio in human evaluations on generated instances;
- a multi-stage evaluation system that evaluates both retrieval and generation performance, result in a comprehensive evaluation on the RAG pipeline;
- robust evaluation metrics derived from rule-based and LLM-based ones, enhancing the reliability of assessments through manual annotations and supervised fine-tuning of an LLM evaluator.
Useful Links: 📝 Paper • 🤗 Hugging Face • 🧩 Github
We have trained two models from Qwen2.5-7B by the lora strategy and human-annotation labels to implement model-based evaluation.Note that the evaluator of hallucination is different from other four.
We provide the evaluator for other metrics except hallucination in this repo.