# Image Classification on Waterbirds (test) (Bias Severity Accuracy) Source: [Image Classification on Waterbirds (test) (Bias Severity Accuracy)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:36:03.422Z Evaluation of Image Classification performance on the test split of the Waterbirds dataset, specifically reporting Average Accuracy metrics across various bias severity thresholds (0.5% to 5.0%). 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 | Image Classification | | Dataset (including any stated split) | Waterbirds (test) | | Metric | Avg Acc (0.5% Bias) | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 10 of 10 | | 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 - [Avg Acc (0.5% Bias)](): Higher is better; unit: PERCENT. - [Avg Acc (1.0% Bias)](): Higher is better; unit: PERCENT. - [Avg Acc (2.0% Bias)](): Higher is better; unit: PERCENT. - [Avg Acc (5.0% Bias)](): Higher is better; unit: PERCENT. ## Recorded results | Method | Avg Acc (0.5% Bias) | Result date | Report paper | | --- | --- | --- | --- | | Intrinsic Feature Guidance | 63.64 | Not recorded | 68fa73b21e209131cea5df8e | | LfF+BE | 61.22 | Not recorded | 68fa73b21e209131cea5df8e | | LfF | 60.66 | Not recorded | 68fa73b21e209131cea5df8e | | DisEnt | 59.59 | Not recorded | 68fa73b21e209131cea5df8e | | LNL | 58.49 | Not recorded | 68fa73b21e209131cea5df8e | | EnD | 58.47 | Not recorded | 68fa73b21e209131cea5df8e | | HEX | 57.88 | Not recorded | 68fa73b21e209131cea5df8e | | Vanilla | 57.41 | Not recorded | 68fa73b21e209131cea5df8e | | ReBias | 55.44 | Not recorded | 68fa73b21e209131cea5df8e | | DisEnt+BE | 51.65 | Not recorded | 68fa73b21e209131cea5df8e | ### Intrinsic Feature Guidance Avg Acc (0.5% Bias): 63.64. Result ID: 699309e53a6c3148491340d6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### LfF+BE Avg Acc (0.5% Bias): 61.22. Result ID: 699309e53a6c3148491340d4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### LfF Avg Acc (0.5% Bias): 60.66. Result ID: 699309e53a6c3148491340d2. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### DisEnt Avg Acc (0.5% Bias): 59.59. Result ID: 699309e53a6c3148491340d3. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### LNL Avg Acc (0.5% Bias): 58.49. Result ID: 699309e53a6c3148491340cf. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### EnD Avg Acc (0.5% Bias): 58.47. Result ID: 699309e53a6c3148491340d0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### HEX Avg Acc (0.5% Bias): 57.88. Result ID: 699309e53a6c3148491340ce. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Vanilla Avg Acc (0.5% Bias): 57.41. Result ID: 699309e53a6c3148491340cd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### ReBias Avg Acc (0.5% Bias): 55.44. Result ID: 699309e53a6c3148491340d1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### DisEnt+BE Avg Acc (0.5% Bias): 51.65. Result ID: 699309e53a6c3148491340d5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.