Skip to content

Latest commit

 

History

History
22 lines (21 loc) · 1.52 KB

rag_eval_multiple_domains_summary_zh.md

File metadata and controls

22 lines (21 loc) · 1.52 KB

RAG Evaluations in LlamaIndex

Multiple Domains Scenarios in ["zh"]

Embedding Models WithoutReranker
[hit_rate/mrr]
CohereRerank
[hit_rate/mrr]
bge-reranker-large
[hit_rate/mrr]
bge-reranker-v2-m3
[hit_rate/mrr]
bce-reranker-base_v1
[hit_rate/mrr]
OpenAI-ada-2 77.35/56.19 85.36/68.13 86.19/69.59 86.74/70.94 88.67/75.26
OpenAI-embed-3-small 84.53/62.51 90.06/70.70 89.78/71.83 91.16/73.72 92.54/78.36
OpenAI-embed-3-large 84.25/60.10 88.12/69.97 89.23/71.21 91.16/73.26 92.27/77.28
bge-large-zh-v1.5 84.81/61.90 89.50/69.81 89.50/71.07 90.61/73.11 92.82/77.64
bge-m3-large 87.57/65.01 91.16/71.39 91.44/72.03 93.09/74.02 94.75/79.46
CohereV3-multilingual 82.87/63.22 86.19/68.14 85.64/68.26 86.74/70.25 88.40/74.42
JinaAI-v2-Base-zh 78.45/56.80 84.81/67.33 84.81/68.14 85.64/69.92 88.12/75.09
gte-large-zh 77.35/55.33 85.36/67.06 85.08/68.83 85.91/70.08 87.85/74.79
e5-large-multilingual 87.02/65.26 89.78/70.73 90.33/71.51 90.88/73.74 92.82/78.69
bce-embedding-base_v1 83.70/62.90 92.27/71.94 92.27/72.79 92.54/75.11 95.03/79.57