# Classification and Retrieval on DataComp small 38 downstream tasks Source: [Classification and Retrieval on DataComp small 38 downstream tasks]() Collection index: [llms.txt]() Data updated: 2026-06-02T09:58:13.624Z Evaluation of combined Classification and Retrieval performance across the 38 downstream tasks included in the small DataComp benchmark suite. 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 | Classification and Retrieval | | Dataset (including any stated split) | DataComp small 38 downstream tasks | | Metric | Mean Score | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 12 of 12 | | 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 - [Mean Score](): Higher is better; unit: PERCENT. ## Recorded results | Method | Mean Score | Result date | Report paper | | --- | --- | --- | --- | | CLIPScore | 17.3 | Not recorded | 68fa79af1e209131cea7edfc | | Basic | 14.2 | Not recorded | 68fa79af1e209131cea7edfc | | DKLR | 13.44 | Not recorded | 68fa79af1e209131cea7edfc | | DKL | 13.37 | Not recorded | 68fa79af1e209131cea7edfc | | LAION2B | 13.3 | Not recorded | 68fa79af1e209131cea7edfc | | No filtering | 13.2 | Not recorded | 68fa79af1e209131cea7edfc | | DC | 13.19 | Not recorded | 68fa79af1e209131cea7edfc | | DW | 12.46 | Not recorded | 68fa79af1e209131cea7edfc | | DKL | 12.2 | Not recorded | 68fa79af1e209131cea7edfc | | DKLR | 11.6 | Not recorded | 68fa79af1e209131cea7edfc | | DW | 11.04 | Not recorded | 68fa79af1e209131cea7edfc | | DC | 9.94 | Not recorded | 68fa79af1e209131cea7edfc | ### CLIPScore Mean Score: 17.3. Result ID: 6a1ea91ffbc373592c194f1f. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Modal":"N/A"}. This record does not assert independent reproduction. ### Basic Mean Score: 14.2. Result ID: 6a1ea91ffbc373592c194f1e. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Modal":"N/A"}. This record does not assert independent reproduction. ### DKLR Mean Score: 13.44. Result ID: 6a1ea91ffbc373592c194f22. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Modal":"Text"}. This record does not assert independent reproduction. ### DKL Mean Score: 13.37. Result ID: 6a1ea91ffbc373592c194f20. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Modal":"Text"}. This record does not assert independent reproduction. ### LAION2B Mean Score: 13.3. Result ID: 6a1ea91ffbc373592c194f1d. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Modal":"N/A"}. This record does not assert independent reproduction. ### No filtering Mean Score: 13.2. Result ID: 6a1ea91ffbc373592c194f1c. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Modal":"N/A"}. This record does not assert independent reproduction. ### DC Mean Score: 13.19. Result ID: 6a1ea91ffbc373592c194f26. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Modal":"Text"}. This record does not assert independent reproduction. ### DW Mean Score: 12.46. Result ID: 6a1ea91ffbc373592c194f24. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Modal":"Text"}. This record does not assert independent reproduction. ### DKL Mean Score: 12.2. Result ID: 6a1ea91ffbc373592c194f21. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Modal":"Image"}. This record does not assert independent reproduction. ### DKLR Mean Score: 11.6. Result ID: 6a1ea91ffbc373592c194f23. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Modal":"Image"}. This record does not assert independent reproduction. ### DW Mean Score: 11.04. Result ID: 6a1ea91ffbc373592c194f25. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Modal":"Image"}. This record does not assert independent reproduction. ### DC Mean Score: 9.94. Result ID: 6a1ea91ffbc373592c194f27. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Modal":"Image"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.