Model card
ColBERTv2.
Stanfordopen-source110M paramsBERT (late interaction)
ColBERTv2: Effective and Efficient Retrieval via Lightweight Late Interaction. NAACL 2022.
§ 01 · Benchmarks
Every benchmark ColBERTv2 has a recorded score for.
| # | Benchmark | Area · Task | Metric | Value | Rank | Date | Source |
|---|---|---|---|---|---|---|---|
| 01 | MS MARCO | Natural Language Processing · Text Ranking | mrr@10 | 39.7% | #3 | 2022-07-10 | source ↗ |
| 02 | BEIR | Natural Language Processing · Text Ranking | ndcg@10 | 49.4% | #4 | 2022-07-10 | source ↗ |
Rank column shows this model’s position vs all other models scored on the same benchmark + metric (competitors after the slash). #1 in red means current SOTA. Sorted by rank, then newest result.
§ 02 · Strengths by area
Where ColBERTv2 actually performs.
§ 03 · Papers
1 paper with results for ColBERTv2.
- 2022-07-10· Natural Language Processing· 2 results
ColBERTv2: Effective and Efficient Retrieval via Lightweight Late Interaction
§ 04 · Related models
Other Stanford models scored on Codesota.
§ 05 · Sources & freshness
Where these numbers come from.
arxiv
2
results
2 of 2 rows marked verified.