# Information Retrieval on MS MARCO TREC Deep Learning

Source: [Information Retrieval on MS MARCO TREC Deep Learning](<https://www.sota2.com/research/sota/information-retrieval-on-ms-marco-trec-deep-learning>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-06-15T06:59:11.245Z

Evaluation of Information Retrieval performance on the MS MARCO TREC Deep Learning challenge data, reporting results for DL19, DL20, and DLH metrics.

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

| Scope | Recorded value |
| --- | --- |
| Task | Information Retrieval |
| Dataset (including any stated split) | MS MARCO TREC Deep Learning |
| Metric | DL19 Score |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 18 of 18 |
| Coverage | All results currently returned for this benchmark |

Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol.

## Metrics

- [DL19 Score](<https://www.sota2.com/research/sota/information-retrieval-on-ms-marco-trec-deep-learning.md?metric=DL19%20Score>): Higher is better; unit: SCALAR.

- [DL20 Score](<https://www.sota2.com/research/sota/information-retrieval-on-ms-marco-trec-deep-learning.md?metric=DL20%20Score>): Higher is better; unit: SCALAR.

- [DLH Score](<https://www.sota2.com/research/sota/information-retrieval-on-ms-marco-trec-deep-learning.md?metric=DLH%20Score>): Higher is better; unit: SCALAR.

## Recorded results

| Method | DL19 Score | Result date | Report paper |
| --- | --- | --- | --- |
| ADORE | 0.713 | Not recorded | 6a2f8719d33dfb23e783bfe8 |
| BGE-base-en-v1.5 | 0.702 | Not recorded | 6a2f8719d33dfb23e783bfe8 |
| ThinkQE | 0.697 | Not recorded | 6a2f8719d33dfb23e783bfe8 |
| MUGI | 0.695 | Not recorded | 6a2f8719d33dfb23e783bfe8 |
| Query2Doc (FS) | 0.69 | Not recorded | 6a2f8719d33dfb23e783bfe8 |
| CSQE | 0.69 | Not recorded | 6a2f8719d33dfb23e783bfe8 |
| Query2Doc (ZS) | 0.687 | Not recorded | 6a2f8719d33dfb23e783bfe8 |
| QA-Expand | 0.683 | Not recorded | 6a2f8719d33dfb23e783bfe8 |
| Query2Doc (CoT) | 0.653 | Not recorded | 6a2f8719d33dfb23e783bfe8 |
| ANCE | 0.645 | Not recorded | 6a2f8719d33dfb23e783bfe8 |
| LameR | 0.637 | Not recorded | 6a2f8719d33dfb23e783bfe8 |
| DPR | 0.622 | Not recorded | 6a2f8719d33dfb23e783bfe8 |
| Contriever-FT | 0.621 | Not recorded | 6a2f8719d33dfb23e783bfe8 |
| Query2E | 0.594 | Not recorded | 6a2f8719d33dfb23e783bfe8 |
| GenQREnsemble | 0.559 | Not recorded | 6a2f8719d33dfb23e783bfe8 |
| GenQR | 0.548 | Not recorded | 6a2f8719d33dfb23e783bfe8 |
| BM25 + RM3 | 0.515 | Not recorded | 6a2f8719d33dfb23e783bfe8 |
| BM25 | 0.506 | Not recorded | 6a2f8719d33dfb23e783bfe8 |

### ADORE

DL19 Score: 0.713. Result ID: 6a2fa2a9fbc373592c2b62ce.

[Reporting paper](<https://www.sota2.com/research/paper/6a2f8719d33dfb23e783bfe8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a2f8719d33dfb23e783bfe8.md>)

Recorded evaluation configuration: {"Category":"Query Expansions w/ BM25","Backbone LLM":"GPT-4.1"}. This record does not assert independent reproduction.

### BGE-base-en-v1.5

DL19 Score: 0.702. Result ID: 6a2fa2a9fbc373592c2b62c2.

[Reporting paper](<https://www.sota2.com/research/paper/6a2f8719d33dfb23e783bfe8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74e31e209131cea64745.md>)

Recorded evaluation configuration: {"Category":"Dense"}. This record does not assert independent reproduction.

### ThinkQE

DL19 Score: 0.697. Result ID: 6a2fa2a9fbc373592c2b62cd.

[Reporting paper](<https://www.sota2.com/research/paper/6a2f8719d33dfb23e783bfe8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79631e209131cea7d5a5.md>)

Recorded evaluation configuration: {"Category":"Query Expansions w/ BM25","Backbone LLM":"GPT-4.1"}. This record does not assert independent reproduction.

### MUGI

DL19 Score: 0.695. Result ID: 6a2fa2a9fbc373592c2b62cb.

[Reporting paper](<https://www.sota2.com/research/paper/6a2f8719d33dfb23e783bfe8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73ef1e209131cea5ef6d.md>)

Recorded evaluation configuration: {"Category":"Query Expansions w/ BM25","Backbone LLM":"GPT-4.1"}. This record does not assert independent reproduction.

### Query2Doc (FS)

DL19 Score: 0.69. Result ID: 6a2fa2a9fbc373592c2b62c8.

[Reporting paper](<https://www.sota2.com/research/paper/6a2f8719d33dfb23e783bfe8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70d71e209131cea4d93f.md>)

Recorded evaluation configuration: {"Category":"Query Expansions w/ BM25","Backbone LLM":"GPT-4.1","Mode":"Few-Shot"}. This record does not assert independent reproduction.

### CSQE

DL19 Score: 0.69. Result ID: 6a2fa2a9fbc373592c2b62cc.

[Reporting paper](<https://www.sota2.com/research/paper/6a2f8719d33dfb23e783bfe8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72541e209131cea55af7.md>)

Recorded evaluation configuration: {"Category":"Query Expansions w/ BM25","Backbone LLM":"GPT-4.1"}. This record does not assert independent reproduction.

### Query2Doc (ZS)

DL19 Score: 0.687. Result ID: 6a2fa2a9fbc373592c2b62c9.

[Reporting paper](<https://www.sota2.com/research/paper/6a2f8719d33dfb23e783bfe8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70d71e209131cea4d93f.md>)

Recorded evaluation configuration: {"Category":"Query Expansions w/ BM25","Backbone LLM":"GPT-4.1","Mode":"Zero-Shot"}. This record does not assert independent reproduction.

### QA-Expand

DL19 Score: 0.683. Result ID: 6a2fa2a9fbc373592c2b62c6.

[Reporting paper](<https://www.sota2.com/research/paper/6a2f8719d33dfb23e783bfe8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7ab11e209131cea84a6e.md>)

Recorded evaluation configuration: {"Category":"Query Expansions w/ BM25","Backbone LLM":"GPT-4.1"}. This record does not assert independent reproduction.

### Query2Doc (CoT)

DL19 Score: 0.653. Result ID: 6a2fa2a9fbc373592c2b62c7.

[Reporting paper](<https://www.sota2.com/research/paper/6a2f8719d33dfb23e783bfe8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70d71e209131cea4d93f.md>)

Recorded evaluation configuration: {"Category":"Query Expansions w/ BM25","Backbone LLM":"GPT-4.1","Mode":"CoT"}. This record does not assert independent reproduction.

### ANCE

DL19 Score: 0.645. Result ID: 6a2fa2a9fbc373592c2b62c0.

[Reporting paper](<https://www.sota2.com/research/paper/6a2f8719d33dfb23e783bfe8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa68881e209131cea1cb62.md>)

[Implementation](<https://aka.ms/ance>)

Recorded evaluation configuration: {"Category":"Dense"}. This record does not assert independent reproduction.

### LameR

DL19 Score: 0.637. Result ID: 6a2fa2a9fbc373592c2b62ca.

[Reporting paper](<https://www.sota2.com/research/paper/6a2f8719d33dfb23e783bfe8.md>)

Recorded evaluation configuration: {"Category":"Query Expansions w/ BM25","Backbone LLM":"GPT-4.1"}. This record does not assert independent reproduction.

### DPR

DL19 Score: 0.622. Result ID: 6a2fa2a9fbc373592c2b62bf.

[Reporting paper](<https://www.sota2.com/research/paper/6a2f8719d33dfb23e783bfe8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa686d1e209131cea1c190.md>)

[Implementation](<https://github.com/facebookresearch/DPR>)

Recorded evaluation configuration: {"Category":"Dense"}. This record does not assert independent reproduction.

### Contriever-FT

DL19 Score: 0.621. Result ID: 6a2fa2a9fbc373592c2b62c1.

[Reporting paper](<https://www.sota2.com/research/paper/6a2f8719d33dfb23e783bfe8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d4b1e209131cea38d50.md>)

[Implementation](<https://github.com/facebookresearch/contriever>)

Recorded evaluation configuration: {"Category":"Dense"}. This record does not assert independent reproduction.

### Query2E

DL19 Score: 0.594. Result ID: 6a2fa2a9fbc373592c2b62c5.

[Reporting paper](<https://www.sota2.com/research/paper/6a2f8719d33dfb23e783bfe8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f491e209131cea44ca0.md>)

Recorded evaluation configuration: {"Category":"Query Expansions w/ BM25","Backbone LLM":"GPT-4.1"}. This record does not assert independent reproduction.

### GenQREnsemble

DL19 Score: 0.559. Result ID: 6a2fa2a9fbc373592c2b62c4.

[Reporting paper](<https://www.sota2.com/research/paper/6a2f8719d33dfb23e783bfe8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72d41e209131cea58c67.md>)

Recorded evaluation configuration: {"Category":"Query Expansions w/ BM25","Backbone LLM":"GPT-4.1"}. This record does not assert independent reproduction.

### GenQR

DL19 Score: 0.548. Result ID: 6a2fa2a9fbc373592c2b62c3.

[Reporting paper](<https://www.sota2.com/research/paper/6a2f8719d33dfb23e783bfe8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa703c1e209131cea49d67.md>)

Recorded evaluation configuration: {"Category":"Query Expansions w/ BM25","Backbone LLM":"GPT-4.1"}. This record does not assert independent reproduction.

### BM25 + RM3

DL19 Score: 0.515. Result ID: 6a2fa2a9fbc373592c2b62be.

[Reporting paper](<https://www.sota2.com/research/paper/6a2f8719d33dfb23e783bfe8.md>)

Recorded evaluation configuration: {"Category":"Sparse"}. This record does not assert independent reproduction.

### BM25

DL19 Score: 0.506. Result ID: 6a2fa2a9fbc373592c2b62bd.

[Reporting paper](<https://www.sota2.com/research/paper/6a2f8719d33dfb23e783bfe8.md>)

Recorded evaluation configuration: {"Category":"Sparse"}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.