# Information Retrieval on MS MARCO in-domain

Source: [Information Retrieval on MS MARCO in-domain](<https://www.sota2.com/research/sota/information-retrieval-on-ms-marco-in-domain>)

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

Data updated: 2026-02-26T05:53:05.222Z

Evaluation of Information Retrieval performance on the in-domain split of the MS MARCO dataset, using metrics such as NDCG@10, MRR@10, and Recall@1000.

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 in-domain |
| Metric | NDCG@10 |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 19 of 19 |
| 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

- [NDCG@10](<https://www.sota2.com/research/sota/information-retrieval-on-ms-marco-in-domain.md?metric=NDCG%4010>): Higher is better; unit: SCALAR.

- [MRR@10](<https://www.sota2.com/research/sota/information-retrieval-on-ms-marco-in-domain.md?metric=MRR%4010>): Higher is better; unit: SCALAR.

- [Recall@1000](<https://www.sota2.com/research/sota/information-retrieval-on-ms-marco-in-domain.md?metric=Recall%401000>): Higher is better; unit: PERCENT.

## Recorded results

| Method | NDCG@10 | Result date | Report paper |
| --- | --- | --- | --- |
| T5-ColBERT | 0.473 | Not recorded | 68fa72ce1e209131cea589a7 |
| XTR | 0.466 | Not recorded | 68fa72ce1e209131cea589a7 |
| ColBERT v2 | 0.463 | Not recorded | 68fa72ce1e209131cea589a7 |
| T5-ColBERT | 0.456 | Not recorded | 68fa72ce1e209131cea589a7 |
| XTR | 0.45 | Not recorded | 68fa72ce1e209131cea589a7 |
| GTR | 0.442 | Not recorded | 68fa6b671e209131cea2e15a |
| GTR | 0.442 | Not recorded | 68fa72ce1e209131cea589a7 |
| GTR | 0.439 | Not recorded | 68fa6b671e209131cea2e15a |
| Splade v2 | 0.433 | Not recorded | 68fa72ce1e209131cea589a7 |
| GTR | 0.43 | Not recorded | 68fa6b671e209131cea2e15a |
| GTR | 0.42 | Not recorded | 68fa6b671e209131cea2e15a |
| GTR | 0.42 | Not recorded | 68fa72ce1e209131cea589a7 |
| TAS-Balanced | 0.408 | Not recorded | 68fa6b671e209131cea2e15a |
| GenQ | 0.408 | Not recorded | 68fa72ce1e209131cea589a7 |
| ColBERT | 0.401 | Not recorded | 68fa6b671e209131cea2e15a |
| ColBERT | 0.401 | Not recorded | 68fa72ce1e209131cea589a7 |
| ANCE | 0.388 | Not recorded | 68fa6b671e209131cea2e15a |
| BM25 | 0.228 | Not recorded | 68fa72ce1e209131cea589a7 |

### T5-ColBERT

NDCG@10: 0.473. Result ID: 69927c543a6c3148490f49e2.

[Reporting paper](<https://www.sota2.com/research/paper/68fa72ce1e209131cea589a7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ddb1e209131cea3c7a7.md>)

Recorded evaluation configuration: {"Retriever Deployment":"One Retriever for All","Model Size":"xxl"}. This record does not assert independent reproduction.

### XTR

NDCG@10: 0.466. Result ID: 69927c543a6c3148490f49e3.

[Reporting paper](<https://www.sota2.com/research/paper/68fa72ce1e209131cea589a7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72ce1e209131cea589a7.md>)

Recorded evaluation configuration: {"Retriever Deployment":"One Retriever for All","Model Size":"xxl"}. This record does not assert independent reproduction.

### ColBERT v2

NDCG@10: 0.463. Result ID: 69927c543a6c3148490f49e0.

[Reporting paper](<https://www.sota2.com/research/paper/68fa72ce1e209131cea589a7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6cf41e209131cea36cce.md>)

[Implementation](<https://github.com/stanford-futuredata/ColBERT>)

Recorded evaluation configuration: {"Retriever Deployment":"One Retriever for All","model-based hard negatives":true}. This record does not assert independent reproduction.

### T5-ColBERT

NDCG@10: 0.456. Result ID: 69927c543a6c3148490f49dd.

[Reporting paper](<https://www.sota2.com/research/paper/68fa72ce1e209131cea589a7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ddb1e209131cea3c7a7.md>)

Recorded evaluation configuration: {"Retriever Deployment":"One Retriever for All","Model Size":"base"}. This record does not assert independent reproduction.

### XTR

NDCG@10: 0.45. Result ID: 69927c543a6c3148490f49de.

[Reporting paper](<https://www.sota2.com/research/paper/68fa72ce1e209131cea589a7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72ce1e209131cea589a7.md>)

Recorded evaluation configuration: {"Retriever Deployment":"One Retriever for All","Model Size":"base"}. This record does not assert independent reproduction.

### GTR

NDCG@10: 0.442. Result ID: 698e883cffc3901de09e691b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6b671e209131cea2e15a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b671e209131cea2e15a.md>)

[Implementation](<https://tfhub.dev/google/collections/gtr/1>)

Recorded evaluation configuration: {"Model Size":"XXL"}. This record does not assert independent reproduction.

### GTR

NDCG@10: 0.442. Result ID: 69927c543a6c3148490f49e1.

[Reporting paper](<https://www.sota2.com/research/paper/68fa72ce1e209131cea589a7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b671e209131cea2e15a.md>)

[Implementation](<https://tfhub.dev/google/collections/gtr/1>)

Recorded evaluation configuration: {"Retriever Deployment":"One Retriever for All","Model Size":"xxl"}. This record does not assert independent reproduction.

### GTR

NDCG@10: 0.439. Result ID: 698e883cffc3901de09e691a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6b671e209131cea2e15a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b671e209131cea2e15a.md>)

[Implementation](<https://tfhub.dev/google/collections/gtr/1>)

Recorded evaluation configuration: {"Model Size":"XL"}. This record does not assert independent reproduction.

### Splade v2

NDCG@10: 0.433. Result ID: 69927c543a6c3148490f49df.

[Reporting paper](<https://www.sota2.com/research/paper/68fa72ce1e209131cea589a7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ad91e209131cea2ac3e.md>)

[Implementation](<https://github.com/naver/splade>)

Recorded evaluation configuration: {"Retriever Deployment":"One Retriever for All","model-based hard negatives":true}. This record does not assert independent reproduction.

### GTR

NDCG@10: 0.43. Result ID: 698e883cffc3901de09e6919.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6b671e209131cea2e15a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b671e209131cea2e15a.md>)

[Implementation](<https://tfhub.dev/google/collections/gtr/1>)

Recorded evaluation configuration: {"Model Size":"Large"}. This record does not assert independent reproduction.

### GTR

NDCG@10: 0.42. Result ID: 698e883cffc3901de09e6918.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6b671e209131cea2e15a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b671e209131cea2e15a.md>)

[Implementation](<https://tfhub.dev/google/collections/gtr/1>)

Recorded evaluation configuration: {"Model Size":"Base"}. This record does not assert independent reproduction.

### GTR

NDCG@10: 0.42. Result ID: 69927c543a6c3148490f49dc.

[Reporting paper](<https://www.sota2.com/research/paper/68fa72ce1e209131cea589a7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b671e209131cea2e15a.md>)

[Implementation](<https://tfhub.dev/google/collections/gtr/1>)

Recorded evaluation configuration: {"Retriever Deployment":"One Retriever for All","Model Size":"base"}. This record does not assert independent reproduction.

### TAS-Balanced

NDCG@10: 0.408. Result ID: 698e883cffc3901de09e6915.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6b671e209131cea2e15a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a011e209131cea2553e.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### GenQ

NDCG@10: 0.408. Result ID: 69927c543a6c3148490f49d9.

[Reporting paper](<https://www.sota2.com/research/paper/68fa72ce1e209131cea589a7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b0d1e209131cea2bbcb.md>)

[Implementation](<https://github.com/UKPLab/beir>)

Recorded evaluation configuration: {"Retriever Deployment":"One Retriever per Domain"}. This record does not assert independent reproduction.

### ColBERT

NDCG@10: 0.401. Result ID: 698e883cffc3901de09e6916.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6b671e209131cea2e15a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67a71e209131cea1780b.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### ColBERT

NDCG@10: 0.401. Result ID: 69927c543a6c3148490f49db.

[Reporting paper](<https://www.sota2.com/research/paper/68fa72ce1e209131cea589a7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67a71e209131cea1780b.md>)

Recorded evaluation configuration: {"Retriever Deployment":"One Retriever for All"}. This record does not assert independent reproduction.

### ANCE

NDCG@10: 0.388. Result ID: 698e883cffc3901de09e6914.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6b671e209131cea2e15a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b0d1e209131cea2bbcb.md>)

[Implementation](<https://github.com/UKPLab/beir>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### BM25

NDCG@10: 0.228. Result ID: 69927c543a6c3148490f49da.

[Reporting paper](<https://www.sota2.com/research/paper/68fa72ce1e209131cea589a7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6de91e209131cea3cdff.md>)

[Implementation](<https://github.com/facebookresearch/dpr-scale/tree/main/spar>)

Recorded evaluation configuration: {"Retriever Deployment":"One Retriever for All"}. This record does not assert independent reproduction.

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