Hugging Face's logo

Datasets:
hakari-bench
/
NanoLongEmbed

Modalities:
Text
Formats:
parquet
Languages:
English
Size:
1K - 10K
Tags:
Long Context
retrieval
nano
information-retrieval
bm25
dense-retrieval
Libraries:
Datasets
pandas
Polars
Dataset card Data Studio Files
xet
 Community
NanoLongEmbed / README.md
hotchpotch's picture
hotchpotch
Update dataset README for reranking_hybrid candidates
ec27ccc verified
preview code
|
Raw
History Blame Contribute Delete
11.2 kB
metadata 
configs:
  - config_name: corpus
    data_files:
      - split: Nano2WikiMultihopQA
        path: corpus/Nano2WikiMultihopQA-00000-of-00001.parquet
      - split: NanoNarrativeQA
        path: corpus/NanoNarrativeQA-00000-of-00001.parquet
      - split: NanoNeedle
        path: corpus/NanoNeedle-00000-of-00001.parquet
      - split: NanoPasskey
        path: corpus/NanoPasskey-00000-of-00001.parquet
      - split: NanoQMSum
        path: corpus/NanoQMSum-00000-of-00001.parquet
      - split: NanoSummScreenFD
        path: corpus/NanoSummScreenFD-00000-of-00001.parquet
  - config_name: queries
    data_files:
      - split: Nano2WikiMultihopQA
        path: queries/Nano2WikiMultihopQA-00000-of-00001.parquet
      - split: NanoNarrativeQA
        path: queries/NanoNarrativeQA-00000-of-00001.parquet
      - split: NanoNeedle
        path: queries/NanoNeedle-00000-of-00001.parquet
      - split: NanoPasskey
        path: queries/NanoPasskey-00000-of-00001.parquet
      - split: NanoQMSum
        path: queries/NanoQMSum-00000-of-00001.parquet
      - split: NanoSummScreenFD
        path: queries/NanoSummScreenFD-00000-of-00001.parquet
    default: true
  - config_name: qrels
    data_files:
      - split: Nano2WikiMultihopQA
        path: qrels/Nano2WikiMultihopQA-00000-of-00001.parquet
      - split: NanoNarrativeQA
        path: qrels/NanoNarrativeQA-00000-of-00001.parquet
      - split: NanoNeedle
        path: qrels/NanoNeedle-00000-of-00001.parquet
      - split: NanoPasskey
        path: qrels/NanoPasskey-00000-of-00001.parquet
      - split: NanoQMSum
        path: qrels/NanoQMSum-00000-of-00001.parquet
      - split: NanoSummScreenFD
        path: qrels/NanoSummScreenFD-00000-of-00001.parquet
  - config_name: bm25
    data_files:
      - split: Nano2WikiMultihopQA
        path: bm25/Nano2WikiMultihopQA-00000-of-00001.parquet
      - split: NanoNarrativeQA
        path: bm25/NanoNarrativeQA-00000-of-00001.parquet
      - split: NanoNeedle
        path: bm25/NanoNeedle-00000-of-00001.parquet
      - split: NanoPasskey
        path: bm25/NanoPasskey-00000-of-00001.parquet
      - split: NanoQMSum
        path: bm25/NanoQMSum-00000-of-00001.parquet
      - split: NanoSummScreenFD
        path: bm25/NanoSummScreenFD-00000-of-00001.parquet
  - config_name: harrier_oss_v1_270m
    data_files:
      - split: Nano2WikiMultihopQA
        path: harrier_oss_v1_270m/Nano2WikiMultihopQA-00000-of-00001.parquet
      - split: NanoNarrativeQA
        path: harrier_oss_v1_270m/NanoNarrativeQA-00000-of-00001.parquet
      - split: NanoNeedle
        path: harrier_oss_v1_270m/NanoNeedle-00000-of-00001.parquet
      - split: NanoPasskey
        path: harrier_oss_v1_270m/NanoPasskey-00000-of-00001.parquet
      - split: NanoQMSum
        path: harrier_oss_v1_270m/NanoQMSum-00000-of-00001.parquet
      - split: NanoSummScreenFD
        path: harrier_oss_v1_270m/NanoSummScreenFD-00000-of-00001.parquet
  - config_name: reranking_hybrid
    data_files:
      - split: Nano2WikiMultihopQA
        path: reranking_hybrid/Nano2WikiMultihopQA-00000-of-00001.parquet
      - split: NanoNarrativeQA
        path: reranking_hybrid/NanoNarrativeQA-00000-of-00001.parquet
      - split: NanoNeedle
        path: reranking_hybrid/NanoNeedle-00000-of-00001.parquet
      - split: NanoPasskey
        path: reranking_hybrid/NanoPasskey-00000-of-00001.parquet
      - split: NanoQMSum
        path: reranking_hybrid/NanoQMSum-00000-of-00001.parquet
      - split: NanoSummScreenFD
        path: reranking_hybrid/NanoSummScreenFD-00000-of-00001.parquet
language:
  - en
tags:
  - Long Context
  - retrieval
  - nano
  - information-retrieval
  - bm25
  - dense-retrieval
  - reranking
  - hakari-bench
dataset_info:
  - config_name: bm25
    features:
      - name: query-id
        dtype: string
      - name: corpus-ids
        list: string
    splits:
      - name: Nano2WikiMultihopQA
        num_bytes: 641290
        num_examples: 200
      - name: NanoNarrativeQA
        num_bytes: 762290
        num_examples: 200
      - name: NanoNeedle
        num_bytes: 832482
        num_examples: 98
      - name: NanoPasskey
        num_bytes: 837165
        num_examples: 100
      - name: NanoQMSum
        num_bytes: 414690
        num_examples: 200
      - name: NanoSummScreenFD
        num_bytes: 720490
        num_examples: 200
    download_size: 4210841
    dataset_size: 4208407
  - config_name: corpus
    features:
      - name: _id
        dtype: string
      - name: text
        dtype: string
    splits:
      - name: Nano2WikiMultihopQA
        num_bytes: 11283128
        num_examples: 300
      - name: NanoNarrativeQA
        num_bytes: 116191265
        num_examples: 355
      - name: NanoNeedle
        num_bytes: 28226538
        num_examples: 800
      - name: NanoPasskey
        num_bytes: 23182064
        num_examples: 800
      - name: NanoQMSum
        num_bytes: 10515610
        num_examples: 197
      - name: NanoSummScreenFD
        num_bytes: 10373940
        num_examples: 336
    download_size: 95954462
    dataset_size: 199772545
  - config_name: harrier_oss_v1_270m
    features:
      - name: query-id
        dtype: string
      - name: corpus-ids
        list: string
    splits:
      - name: Nano2WikiMultihopQA
        num_bytes: 641290
        num_examples: 200
      - name: NanoNarrativeQA
        num_bytes: 762290
        num_examples: 200
      - name: NanoNeedle
        num_bytes: 842815
        num_examples: 98
      - name: NanoPasskey
        num_bytes: 831334
        num_examples: 100
      - name: NanoQMSum
        num_bytes: 414690
        num_examples: 200
      - name: NanoSummScreenFD
        num_bytes: 720490
        num_examples: 200
    download_size: 4215430
    dataset_size: 4212909
  - config_name: qrels
    features:
      - name: query-id
        dtype: string
      - name: corpus-id
        dtype: string
    splits:
      - name: Nano2WikiMultihopQA
        num_bytes: 4614
        num_examples: 200
      - name: NanoNarrativeQA
        num_bytes: 4616
        num_examples: 200
      - name: NanoNeedle
        num_bytes: 3362
        num_examples: 98
      - name: NanoPasskey
        num_bytes: 3434
        num_examples: 100
      - name: NanoQMSum
        num_bytes: 4576
        num_examples: 200
      - name: NanoSummScreenFD
        num_bytes: 4631
        num_examples: 200
    download_size: 18531
    dataset_size: 25233
  - config_name: queries
    features:
      - name: _id
        dtype: string
      - name: text
        dtype: string
    splits:
      - name: Nano2WikiMultihopQA
        num_bytes: 16812
        num_examples: 200
      - name: NanoNarrativeQA
        num_bytes: 13155
        num_examples: 200
      - name: NanoNeedle
        num_bytes: 7953
        num_examples: 98
      - name: NanoPasskey
        num_bytes: 5997
        num_examples: 100
      - name: NanoQMSum
        num_bytes: 92566
        num_examples: 200
      - name: NanoSummScreenFD
        num_bytes: 123428
        num_examples: 200
    download_size: 165197
    dataset_size: 259911
  - config_name: reranking_hybrid
    features:
      - name: query-id
        dtype: string
      - name: corpus-ids
        list: string
    splits:
      - name: Nano2WikiMultihopQA
        num_bytes: 215691
        num_examples: 200
      - name: NanoNarrativeQA
        num_bytes: 217014
        num_examples: 200
      - name: NanoNeedle
        num_bytes: 170732
        num_examples: 98
      - name: NanoPasskey
        num_bytes: 167789
        num_examples: 100
      - name: NanoQMSum
        num_bytes: 211758
        num_examples: 200
      - name: NanoSummScreenFD
        num_bytes: 216907
        num_examples: 200
    download_size: 1202006
    dataset_size: 1199891

NanoLongEmbed

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

NanoLongEmbed contains Nano-style long-context retrieval splits derived from LongEmbed tasks.

Usage 

from datasets import load_dataset

dataset_id = "hakari-bench/NanoLongEmbed"
split = "Nano2WikiMultihopQA"

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
Nano2WikiMultihopQA 200 300 200 67.5 65.5 79.2 37445.6 31645.5 55443.0
NanoNarrativeQA 200 355 200 49.3 47.5 60.0 326753.0 252633.0 371460.0
NanoNeedle 98 800 98 59.0 60.5 67.0 35246.1 12934.5 44460.5
NanoPasskey 100 800 100 37.8 38.0 39.0 28956.7 10883.0 36355.8
NanoQMSum 200 197 200 446.3 410.5 577.2 53335.8 50063.0 67575.0
NanoSummScreenFD 200 336 200 600.7 507.0 776.2 30854.3 29210.0 37969.8

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 - 82.17 61.91 73.15 97.66 93.40 98.83 - 13
Nano2WikiMultihopQA english_porter_stop 95.03 84.00 91.11 99.00 96.50 100.00 100 0
NanoNarrativeQA english_porter_stop 76.19 33.15 51.20 90.00 75.00 94.50 100-101 11
NanoNeedle english_porter_stop 72.07 60.99 68.23 96.94 95.92 98.98 100-101 1
NanoPasskey english_porter_stop 77.17 64.73 72.94 100.00 100.00 100.00 100 0
NanoQMSum english_porter_stop 74.40 36.60 60.97 100.00 96.00 99.50 100-101 1
NanoSummScreenFD english_porter_stop 98.13 91.98 94.43 100.00 97.00 100.00 100 0

Hybrid Safeguard Summary

  • Safeguard positives: 13
  • Rows limited by corpus size: 0
  • Metadata file: reranking_hybrid_metadata.json

Source Links

License

NanoLongEmbed is a derived dataset. Users must comply with the licenses, terms, and attribution requirements of the upstream datasets listed above.