Datasets › MS MARCO
MS MARCO (Microsoft Machine Reading Comprehension Dataset)
The MS MARCO (Microsoft MAchine Reading Comprehension) is a collection of datasets focused on deep learning in search. The first dataset was a question answering dataset featuring 100,000 real Bing questions and a human generated answer. Over time the collection was extended with a 1,000,000 question dataset, a natural language generation dataset, a passage ranking dataset, keyphrase extraction dataset, crawling dataset, and a conversational search.
Source: https://microsoft.github.io/msmarco/ Image Source: https://arxiv.org/pdf/1809.08267.pdf
Benchmarks archive 2025-07-28
All 7 leaderboards whose dataset resolves to this page shown (sort by any header). "First row" is the archive's own first row at snapshot, in the archive's row order; nothing here re-ranks and metric direction is not asserted.
| First row (archive order) | Paper | Code | ||||
|---|---|---|---|---|---|---|
| Passage Ranking | MS MARCO | Fine-tuned SOTA MRR@10 44.3 | Text and Code Embeddings by Contrastive Pre-Training | openmatch/coco-dr | 4 | Compare |
| Passage Re-Ranking | MS MARCO | HLATR MRR 0.42 | HLATR: Enhance Multi-stage Text Retrieval with Hybrid... | Alibaba-NLP/HLATR | 4 | Compare |
| Question Answering | MS MARCO | Masque Q&A Style Rouge-L 52.2 | Multi-style Generative Reading Comprehension | — | 4 | Compare |
| Information Retrieval | MS MARCO | ConAE-128 Time (ms) 0.3245 | Dimension Reduction for Efficient Dense Retrieval via... | neuir/conae | 3 | Compare |
| Information Retrieval | MSMARCO | RetroMAE MRR@10 0.416 | RetroMAE: Pre-Training Retrieval-oriented Language... | staoxiao/retromae | 1 | Compare |
| Passage Retrieval | MS MARCO | TW-BERT MAP 0.2025 | End-to-End Query Term Weighting | ledmaster/tw-bert | 1 | Compare |
| Text Retrieval | MS MARCO | Lucene (BM25S) NDCG@10 22.8 | BM25S: Orders of magnitude faster lexical search via... | xhluca/bm25s +2 | 1 | Compare |
Papers archive 2025-07-28
14 shown of 14 papers with a leaderboard row on this dataset's benchmarks, newest first. The archive's own "papers using this dataset" list was never published, so this is the benchmark-backed subset; the archive's count for this dataset is 1,036. The Syntology column is from Syntology's graph (read 2026-09-24), stated per sample; it is not part of any archive number.
Dataset loaders archive 2025-07-28
3 loaders as listed in the archive; links are outbound and not re-checked here.
Tasks archive 2025-07-28
License archive 2025-07-28
Custom (research-only, non-commercial)
Modalities archive 2025-07-28
Languages archive 2025-07-28
Variants archive 2025-07-28
- MS MARCO
- MSMARCO
2 variant names, as the archive lists them.
Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections