# Information Retrieval on TREC DL 19 20 (test)

Source: [Information Retrieval on TREC DL 19 20 (test)](<https://www.sota2.com/research/sota/information-retrieval-on-trec-dl-19-20-test>)

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

Data updated: 2026-04-24T08:22:18.536Z

Evaluation of Information Retrieval performance on the combined TREC DL 2019 and TREC DL 2020 test sets, measured using metrics like nDCG@10 and Recall@100.

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) | TREC DL 19 20 (test) |
| Metric | nDCG@10 |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 17 of 17 |
| 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-trec-dl-19-20-test.md?metric=nDCG%4010>): Higher is better; unit: PERCENT.

- [Recall@100](<https://www.sota2.com/research/sota/information-retrieval-on-trec-dl-19-20-test.md?metric=Recall%40100>): Higher is better; unit: PERCENT.

- [QD FLOPs](<https://www.sota2.com/research/sota/information-retrieval-on-trec-dl-19-20-test.md?metric=QD%20FLOPs>): Lower is better; unit: SCALAR.

- [Avg. Doc Length](<https://www.sota2.com/research/sota/information-retrieval-on-trec-dl-19-20-test.md?metric=Avg.%20Doc%20Length>): Higher is better; unit: SCALAR.

- [ΔE²](<https://www.sota2.com/research/sota/information-retrieval-on-trec-dl-19-20-test.md?metric=%CE%94E%C2%B2>): Lower is better; unit: SCALAR.

## Recorded results

| Method | nDCG@10 | Result date | Report paper |
| --- | --- | --- | --- |
| ColBERTv2 | 75.1 | Not recorded | 69eaf93355a3c32af89f8a88 |
| SPLADEv3 | 74.8 | Not recorded | 69eaf93355a3c32af89f8a88 |
| ColBERTv2 | 73.6 | Not recorded | 69eaf93355a3c32af89f8a88 |
| SAE-SPLADE | 72.3 | Not recorded | 69eaf93355a3c32af89f8a88 |
| SAE-SPLADE | 72.1 | Not recorded | 69eaf93355a3c32af89f8a88 |
| SPLADE | 72 | Not recorded | 69eaf93355a3c32af89f8a88 |
| SAE-SPLADE | 71.5 | Not recorded | 69eaf93355a3c32af89f8a88 |
| SimLM | 70.6 | Not recorded | 69eaf93355a3c32af89f8a88 |
| CL-SR | 66 | Not recorded | 69eaf93355a3c32af89f8a88 |
| BM25\* | 64.57 | Not recorded | 6997eab109c33900fcd7671b |
| BM25 | 62.16 | Not recorded | 6997eab109c33900fcd7671b |
| BM25+ | 61.07 | Not recorded | 6997eab109c33900fcd7671b |
| BM25-adpt | 60.9 | Not recorded | 6997eab109c33900fcd7671b |
| QL-Dir\* | 60.68 | Not recorded | 6997eab109c33900fcd7671b |
| QL-Dir | 57.03 | Not recorded | 6997eab109c33900fcd7671b |
| QL-JM | 56.65 | Not recorded | 6997eab109c33900fcd7671b |
| BM25 | 49.2 | Not recorded | 69eaf93355a3c32af89f8a88 |

### ColBERTv2

nDCG@10: 75.1. Result ID: 69eb282747f34ce924d35c4b.

[Reporting paper](<https://www.sota2.com/research/paper/69eaf93355a3c32af89f8a88.md>)

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

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

Recorded evaluation configuration: {"Encoder":"BERTbase"}. This record does not assert independent reproduction.

### SPLADEv3

nDCG@10: 74.8. Result ID: 69eb282747f34ce924d35c4e.

[Reporting paper](<https://www.sota2.com/research/paper/69eaf93355a3c32af89f8a88.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72751e209131cea567ad.md>)

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

Recorded evaluation configuration: {"Encoder":"DistilBERT"}. This record does not assert independent reproduction.

### ColBERTv2

nDCG@10: 73.6. Result ID: 69eb282747f34ce924d35c4a.

[Reporting paper](<https://www.sota2.com/research/paper/69eaf93355a3c32af89f8a88.md>)

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

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

Recorded evaluation configuration: {"Encoder":"DistilBERT"}. This record does not assert independent reproduction.

### SAE-SPLADE

nDCG@10: 72.3. Result ID: 69eb282747f34ce924d35c52.

[Reporting paper](<https://www.sota2.com/research/paper/69eaf93355a3c32af89f8a88.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69eaf93355a3c32af89f8a88.md>)

Recorded evaluation configuration: {"Encoder":"DistilBERT","k":"16"}. This record does not assert independent reproduction.

### SAE-SPLADE

nDCG@10: 72.1. Result ID: 69eb282747f34ce924d35c51.

[Reporting paper](<https://www.sota2.com/research/paper/69eaf93355a3c32af89f8a88.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69eaf93355a3c32af89f8a88.md>)

Recorded evaluation configuration: {"Encoder":"DistilBERT","k":"8"}. This record does not assert independent reproduction.

### SPLADE

nDCG@10: 72. Result ID: 69eb282747f34ce924d35c4f.

[Reporting paper](<https://www.sota2.com/research/paper/69eaf93355a3c32af89f8a88.md>)

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

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

Recorded evaluation configuration: {"Encoder":"DistilBERT"}. This record does not assert independent reproduction.

### SAE-SPLADE

nDCG@10: 71.5. Result ID: 69eb282747f34ce924d35c50.

[Reporting paper](<https://www.sota2.com/research/paper/69eaf93355a3c32af89f8a88.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69eaf93355a3c32af89f8a88.md>)

Recorded evaluation configuration: {"Encoder":"DistilBERT","k":"4"}. This record does not assert independent reproduction.

### SimLM

nDCG@10: 70.6. Result ID: 69eb282747f34ce924d35c49.

[Reporting paper](<https://www.sota2.com/research/paper/69eaf93355a3c32af89f8a88.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f481e209131cea44bd6.md>)

[Implementation](<https://github.com/microsoft/unilm/tree/master/simlm>)

Recorded evaluation configuration: {"Encoder":"BERTbase"}. This record does not assert independent reproduction.

### CL-SR

nDCG@10: 66. Result ID: 69eb282747f34ce924d35c4d.

[Reporting paper](<https://www.sota2.com/research/paper/69eaf93355a3c32af89f8a88.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7b001e209131cea86299.md>)

Recorded evaluation configuration: {"Encoder":"BERTbase"}. This record does not assert independent reproduction.

### BM25\*

nDCG@10: 64.57. Result ID: 699b01a95698665ec493a5d8.

[Reporting paper](<https://www.sota2.com/research/paper/6997eab109c33900fcd7671b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6997eab109c33900fcd7671b.md>)

[Implementation](<https://github.com/fangchenli/ranking-evolved>)

Recorded evaluation configuration: {"evolved":"true"}. This record does not assert independent reproduction.

### BM25

nDCG@10: 62.16. Result ID: 699b01a95698665ec493a5d5.

[Reporting paper](<https://www.sota2.com/research/paper/6997eab109c33900fcd7671b.md>)

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

### BM25+

nDCG@10: 61.07. Result ID: 699b01a95698665ec493a5d6.

[Reporting paper](<https://www.sota2.com/research/paper/6997eab109c33900fcd7671b.md>)

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

### BM25-adpt

nDCG@10: 60.9. Result ID: 699b01a95698665ec493a5d7.

[Reporting paper](<https://www.sota2.com/research/paper/6997eab109c33900fcd7671b.md>)

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

### QL-Dir\*

nDCG@10: 60.68. Result ID: 699b01a95698665ec493a5db.

[Reporting paper](<https://www.sota2.com/research/paper/6997eab109c33900fcd7671b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6997eab109c33900fcd7671b.md>)

[Implementation](<https://github.com/fangchenli/ranking-evolved>)

Recorded evaluation configuration: {"evolved":"true"}. This record does not assert independent reproduction.

### QL-Dir

nDCG@10: 57.03. Result ID: 699b01a95698665ec493a5d9.

[Reporting paper](<https://www.sota2.com/research/paper/6997eab109c33900fcd7671b.md>)

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

### QL-JM

nDCG@10: 56.65. Result ID: 699b01a95698665ec493a5da.

[Reporting paper](<https://www.sota2.com/research/paper/6997eab109c33900fcd7671b.md>)

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

### BM25

nDCG@10: 49.2. Result ID: 69eb282747f34ce924d35c4c.

[Reporting paper](<https://www.sota2.com/research/paper/69eaf93355a3c32af89f8a88.md>)

Recorded evaluation configuration: {"Encoder":"-"}. This record does not assert independent reproduction.

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