nano_bm25_subset_config.json

{
  "dataset": "NanoR2MED",
  "benchmark": "R2MED",
  "query_limit": 200,
  "doc_limit": 10000,
  "bm25_top_k": 100,
  "candidate_policy": "Rank the selected split corpus with BM25, keep up to top-100 candidates, and force qrels-positive documents into the final candidate list when missing.",
  "splits": [
    {
      "split_name": "NanoR2MEDBiology",
      "source_task": "R2MEDBiologyRetrieval",
      "queries": 103,
      "corpus": 10000,
      "qrels": 374,
      "tokenization_plan": {
        "mode": "stemmer",
        "language": "en",
        "stemmer_algorithm": "english",
        "tokenizer_name": null,
        "reason": "language 'en' mapped to stemmer 'english'"
      },
      "ndcg_at_10": 0.25128449065032643,
      "ndcg_at_100": 0.4296835025091202,
      "forced_queries": 68,
      "forced_doc_count": 167,
      "missing_positive_doc_count_after_forcing": 0
    },
    {
      "split_name": "NanoR2MEDBioinformatics",
      "source_task": "R2MEDBioinformaticsRetrieval",
      "queries": 77,
      "corpus": 10000,
      "qrels": 226,
      "tokenization_plan": {
        "mode": "stemmer",
        "language": "en",
        "stemmer_algorithm": "english",
        "tokenizer_name": null,
        "reason": "language 'en' mapped to stemmer 'english'"
      },
      "ndcg_at_10": 0.17857929009122858,
      "ndcg_at_100": 0.35060738065375957,
      "forced_queries": 47,
      "forced_doc_count": 111,
      "missing_positive_doc_count_after_forcing": 0
    },
    {
      "split_name": "NanoR2MEDMedicalSciences",
      "source_task": "R2MEDMedicalSciencesRetrieval",
      "queries": 88,
      "corpus": 10000,
      "qrels": 244,
      "tokenization_plan": {
        "mode": "stemmer",
        "language": "en",
        "stemmer_algorithm": "english",
        "tokenizer_name": null,
        "reason": "language 'en' mapped to stemmer 'english'"
      },
      "ndcg_at_10": 0.10428646688120163,
      "ndcg_at_100": 0.29337732163029967,
      "forced_queries": 62,
      "forced_doc_count": 138,
      "missing_positive_doc_count_after_forcing": 0
    },
    {
      "split_name": "NanoR2MEDMedXpertQAExam",
      "source_task": "R2MEDMedXpertQAExamRetrieval",
      "queries": 97,
      "corpus": 10000,
      "qrels": 292,
      "tokenization_plan": {
        "mode": "stemmer",
        "language": "en",
        "stemmer_algorithm": "english",
        "tokenizer_name": null,
        "reason": "language 'en' mapped to stemmer 'english'"
      },
      "ndcg_at_10": 0.024462598222211914,
      "ndcg_at_100": 0.2304796798768815,
      "forced_queries": 90,
      "forced_doc_count": 255,
      "missing_positive_doc_count_after_forcing": 0
    },
    {
      "split_name": "NanoR2MEDMedQADiag",
      "source_task": "R2MEDMedQADiagRetrieval",
      "queries": 118,
      "corpus": 10000,
      "qrels": 522,
      "tokenization_plan": {
        "mode": "stemmer",
        "language": "en",
        "stemmer_algorithm": "english",
        "tokenizer_name": null,
        "reason": "language 'en' mapped to stemmer 'english'"
      },
      "ndcg_at_10": 0.028104212132602936,
      "ndcg_at_100": 0.2642743613844765,
      "forced_queries": 109,
      "forced_doc_count": 440,
      "missing_positive_doc_count_after_forcing": 0
    },
    {
      "split_name": "NanoR2MEDPMCTreatment",
      "source_task": "R2MEDPMCTreatmentRetrieval",
      "queries": 150,
      "corpus": 10000,
      "qrels": 315,
      "tokenization_plan": {
        "mode": "stemmer",
        "language": "en",
        "stemmer_algorithm": "english",
        "tokenizer_name": null,
        "reason": "language 'en' mapped to stemmer 'english'"
      },
      "ndcg_at_10": 0.01798988334744109,
      "ndcg_at_100": 0.2202813875703761,
      "forced_queries": 106,
      "forced_doc_count": 186,
      "missing_positive_doc_count_after_forcing": 0
    },
    {
      "split_name": "NanoR2MEDPMCClinical",
      "source_task": "R2MEDPMCClinicalRetrieval",
      "queries": 114,
      "corpus": 10000,
      "qrels": 248,
      "tokenization_plan": {
        "mode": "stemmer",
        "language": "en",
        "stemmer_algorithm": "english",
        "tokenizer_name": null,
        "reason": "language 'en' mapped to stemmer 'english'"
      },
      "ndcg_at_10": 0.3277008274619314,
      "ndcg_at_100": 0.4621547795535059,
      "forced_queries": 46,
      "forced_doc_count": 63,
      "missing_positive_doc_count_after_forcing": 0
    },
    {
      "split_name": "NanoR2MEDIIYiClinical",
      "source_task": "R2MEDIIYiClinicalRetrieval",
      "queries": 129,
      "corpus": 10000,
      "qrels": 457,
      "tokenization_plan": {
        "mode": "stemmer",
        "language": "en",
        "stemmer_algorithm": "english",
        "tokenizer_name": null,
        "reason": "language 'en' mapped to stemmer 'english'"
      },
      "ndcg_at_10": 0.12460354977311297,
      "ndcg_at_100": 0.33378196865460924,
      "forced_queries": 113,
      "forced_doc_count": 268,
      "missing_positive_doc_count_after_forcing": 0
    }
  ]
}