# Visual Emotion Distribution Learning on Twitter\_LDL (test) Source: [Visual Emotion Distribution Learning on Twitter\_LDL (test)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:39:06.755Z Evaluation of Visual Emotion Distribution Learning performance on the test split of the Twitter\_LDL dataset, reporting metrics including Chebyshev, KL divergence, and Accuracy. 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 | Visual Emotion Distribution Learning | | Dataset (including any stated split) | Twitter\_LDL (test) | | Metric | Chebyshev Distance | | Unit | SCALAR | | Direction | Lower is better | | Results loaded | 14 of 14 | | 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 - [Chebyshev Distance](): Lower is better; unit: SCALAR. - [Clark Distance](): Lower is better; unit: SCALAR. - [Canberra Distance](): Lower is better; unit: SCALAR. - [KL Divergence](): Lower is better; unit: SCALAR. - [Cosine Similarity](): Higher is better; unit: SCALAR. - [Intersection Score](): Higher is better; unit: PERCENT. - [Average Rank](): Lower is better; unit: SCALAR. - [Accuracy](): Higher is better; unit: PERCENT. ## Recorded results | Method | Chebyshev Distance | Result date | Report paper | | --- | --- | --- | --- | | Emotion Circle (Ours) | 0.22 | Not recorded | 68fa6a3b1e209131cea26d4e | | E-GCN | 0.24 | Not recorded | 68fa6a3b1e209131cea26d4e | | JCDL | 0.25 | Not recorded | 68fa6a3b1e209131cea26d4e | | SSDL | 0.25 | Not recorded | 68fa6a3b1e209131cea26d4e | | DLDL | 0.26 | Not recorded | 68fa6a3b1e209131cea26d4e | | ACPNN | 0.27 | Not recorded | 68fa6a3b1e209131cea26d4e | | AA-kNN | 0.28 | Not recorded | 68fa6a3b1e209131cea26d4e | | SA-IIS | 0.28 | Not recorded | 68fa6a3b1e209131cea26d4e | | CNNR | 0.28 | Not recorded | 68fa6a3b1e209131cea26d4e | | SA-CPNN | 0.36 | Not recorded | 68fa6a3b1e209131cea26d4e | | AA-BP | 0.37 | Not recorded | 68fa6a3b1e209131cea26d4e | | SA-BFGS | 0.37 | Not recorded | 68fa6a3b1e209131cea26d4e | | PT-Bayes | 0.53 | Not recorded | 68fa6a3b1e209131cea26d4e | | PT-SVM | 0.63 | Not recorded | 68fa6a3b1e209131cea26d4e | ### Emotion Circle (Ours) Chebyshev Distance: 0.22. Result ID: 698e2109ffc3901de09b0294. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"CNN-based","Backbone":"ResNet-50","Loss Function":"Progressive Circular loss"}. This record does not assert independent reproduction. ### E-GCN Chebyshev Distance: 0.24. Result ID: 698e2109ffc3901de09b0293. [Reporting paper]() Recorded evaluation configuration: {"Category":"CNN-based","Backbone":"ResNet-50"}. This record does not assert independent reproduction. ### JCDL Chebyshev Distance: 0.25. Result ID: 698e2109ffc3901de09b0291. [Reporting paper]() Recorded evaluation configuration: {"Category":"CNN-based","Backbone":"ResNet-50"}. This record does not assert independent reproduction. ### SSDL Chebyshev Distance: 0.25. Result ID: 698e2109ffc3901de09b0292. [Reporting paper]() Recorded evaluation configuration: {"Category":"CNN-based","Backbone":"ResNet-50"}. This record does not assert independent reproduction. ### DLDL Chebyshev Distance: 0.26. Result ID: 698e2109ffc3901de09b028f. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Category":"CNN-based","Backbone":"ResNet-50"}. This record does not assert independent reproduction. ### ACPNN Chebyshev Distance: 0.27. Result ID: 698e2109ffc3901de09b0290. [Reporting paper]() Recorded evaluation configuration: {"Category":"CNN-based","Backbone":"ResNet-50"}. This record does not assert independent reproduction. ### AA-kNN Chebyshev Distance: 0.28. Result ID: 698e2109ffc3901de09b0289. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"Algorithm Adaptation (AA)"}. This record does not assert independent reproduction. ### SA-IIS Chebyshev Distance: 0.28. Result ID: 698e2109ffc3901de09b028b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"Specialized Algorithms (SA)"}. This record does not assert independent reproduction. ### CNNR Chebyshev Distance: 0.28. Result ID: 698e2109ffc3901de09b028e. [Reporting paper]() Recorded evaluation configuration: {"Category":"CNN-based","Backbone":"ResNet-50"}. This record does not assert independent reproduction. ### SA-CPNN Chebyshev Distance: 0.36. Result ID: 698e2109ffc3901de09b028d. [Reporting paper]() Recorded evaluation configuration: {"Category":"Specialized Algorithms (SA)"}. This record does not assert independent reproduction. ### AA-BP Chebyshev Distance: 0.37. Result ID: 698e2109ffc3901de09b028a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"Algorithm Adaptation (AA)"}. This record does not assert independent reproduction. ### SA-BFGS Chebyshev Distance: 0.37. Result ID: 698e2109ffc3901de09b028c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"Specialized Algorithms (SA)"}. This record does not assert independent reproduction. ### PT-Bayes Chebyshev Distance: 0.53. Result ID: 698e2109ffc3901de09b0287. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"Problem Transformation (PT)"}. This record does not assert independent reproduction. ### PT-SVM Chebyshev Distance: 0.63. Result ID: 698e2109ffc3901de09b0288. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"Problem Transformation (PT)"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.