NanoCMTEB is a compact Chinese retrieval benchmark assembled from C-MTEB / MTEB(cmn, v1) retrieval tasks. It covers Chinese medical, web, e-commerce, video, and general question-answering retrieval settings while keeping small Nano splits for fast evaluation.

Usage

from datasets import load_dataset

dataset_id = "hakari-bench/NanoCMTEB"
split = "cmedqa"

queries = load_dataset(dataset_id, "queries", split=split)
corpus = load_dataset(dataset_id, "corpus", split=split)
qrels = load_dataset(dataset_id, "qrels", split=split)
reranking_candidates = load_dataset(dataset_id, "reranking_hybrid", split=split)

Data Layout

This dataset uses six Hugging Face Datasets configs:

corpus: documents with _id and text
queries: queries with _id and text
qrels: positive relevance labels with query-id and corpus-id
bm25: BM25 candidate lists with query-id and corpus-ids
harrier_oss_v1_270m: dense candidate lists from microsoft/harrier-oss-v1-270m
reranking_hybrid: RRF candidate lists built from bm25 and harrier_oss_v1_270m

Each config has the same Nano split names.

Candidate Construction

bm25: local BM25 top-500 with automatic language-aware tokenization. The resolved tokenizer is shown in the Candidate Quality table, for example wordseg@ja.
harrier_oss_v1_270m: dense top-500 from microsoft/harrier-oss-v1-270m. In tables this is shown as Dense; Dense means microsoft/harrier-oss-v1-270m with the web_search_query prompt for queries and cosine similarity over normalized embeddings.
reranking_hybrid: RRF over bm25 and harrier_oss_v1_270m using rrf_k=100, keeping the RRF top-100.

Safeguard means rank 101 is appended only when RRF top-100 contains no qrels-positive document.

Split Statistics

Length statistics are character counts computed with len(str(text)).

Nano split	Queries	Corpus	Qrels	Query chars avg	Query chars p50	Query chars p75	Doc chars avg	Doc chars p50	Doc chars p75
cmedqa	200	10000	324	52.0	44.0	69.0	157.6	112.0	203.0
covid	200	10000	204	25.7	24.0	30.0	409.3	275.0	329.0
du	200	10000	889	9.1	9.0	10.0	397.4	280.0	356.0
ecom	200	10000	200	6.9	6.0	8.0	33.1	31.0	34.0
medical	200	10000	200	18.1	16.0	20.2	119.7	101.0	149.0
mmarco	200	10000	212	10.4	9.0	12.0	113.9	102.0	136.0
t2	200	10000	979	10.7	10.0	13.0	913.5	482.5	983.0
video	200	10000	200	7.1	6.0	9.0	30.5	24.0	30.0

Candidate Quality

nDCG@10 and Recall@100 are computed from the included candidate rankings against the included qrels, then reported as 0-100 scores such as 52.45. Recall@100 uses only the top 100 candidates; an optional rank-101 safeguard positive is not counted in Recall@100.

Dense means microsoft/harrier-oss-v1-270m with the web_search_query prompt and cosine similarity.

Nano split	BM25 tokenizer	BM25 nDCG@10	Dense nDCG@10	Hybrid nDCG@10	BM25 Recall@100	Dense Recall@100	Hybrid Recall@100	Hybrid candidates	Safeguard positives
Mean	-	60.03	75.75	68.84	78.26	91.74	92.14	-	108
cmedqa	wordseg@zh	16.67	33.22	25.95	39.12	72.47	66.12	100-101	56
covid	wordseg@zh	78.88	75.18	78.34	96.00	93.00	99.00	100-101	2
du	wordseg@zh	73.37	92.86	82.24	84.76	97.42	97.69	100-101	2
ecom	wordseg@zh	59.13	80.52	70.25	82.50	95.50	96.00	100-101	8
medical	wordseg@zh	35.82	56.91	46.99	56.00	84.00	82.50	100-101	35
mmarco	wordseg@zh	67.95	88.59	79.84	90.75	97.00	98.50	100-101	3
t2	wordseg@zh	79.44	92.45	86.04	87.44	96.01	97.78	100-101	1
video	wordseg@zh	68.97	86.29	81.03	89.50	98.50	99.50	100-101	1

Hybrid Safeguard Summary

Safeguard positives: 108
Rows limited by corpus size: 0
Metadata file: reranking_hybrid_metadata.json

Source Links

License

NanoCMTEB is a derived dataset. Users must comply with the licenses, terms, and attribution requirements of the upstream MTEB task sources and their original datasets.

Downloads last month: 167

Total file size:

71.6 MB