NanoRuMTEB

This dataset is a Nano-style retrieval dataset for HAKARI-bench.

NanoRuMTEB is a compact Russian retrieval benchmark assembled from ruMTEB retrieval tasks. It includes Russian MIRACL, RIA News retrieval, and RuBQ retrieval; RuSciBench retrieval tasks are grouped separately under NanoMTEB-Misc.

Usage

from datasets import load_dataset

dataset_id = "hakari-bench/NanoRuMTEB"
split = "miracl_ru"

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
miracl_ru	200	10000	579	45.4	41.0	53.2	517.3	404.0	685.2
ria_news	200	10000	200	62.0	65.0	68.0	1145.3	1169.5	1523.0
ru_bq	200	10000	334	52.2	48.0	60.2	484.5	380.0	645.0

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	-	70.89	87.18	78.95	91.83	95.94	98.73	-	5
miracl_ru	stemmer@russian	51.54	79.38	66.46	89.09	95.78	99.26	100-101	1
ria_news	stemmer@russian	91.35	94.78	92.72	97.50	97.50	99.00	100-101	2
ru_bq	stemmer@russian	69.79	87.39	77.67	88.92	94.54	97.92	100-101	2

Hybrid Safeguard Summary

• Safeguard positives: 5
• Rows limited by corpus size: 0
• Metadata file: reranking_hybrid_metadata.json

Source Links

• mteb/MIRACLRetrievalHardNegatives
• mteb/RiaNewsRetrieval_test_top_250_only_w_correct-v2
• ai-forever/rubq-retrieval

License

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