# Polarization Detection on SemEval Task 9 2026 (test) Source: [Polarization Detection on SemEval Task 9 2026 (test)]() Collection index: [llms.txt]() Data updated: 2026-05-07T10:01:58.822Z Evaluation of Polarization Detection performance on the test split of the SemEval Task 9 2026 dataset, primarily measured using Macro-F1 score and tracking against the current State-of-the-Art. 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 | Polarization Detection | | Dataset (including any stated split) | SemEval Task 9 2026 (test) | | Metric | Macro-F1 | | 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 - [Macro-F1](): Higher is better; unit: PERCENT. - [SOTA Macro-F1](): Higher is better; unit: PERCENT. - [Delta SOTA](): Higher is better; unit: SCALAR. - [Rank](): Lower is better; unit: SCALAR. ## Recorded results | Method | Macro-F1 | Result date | Report paper | | --- | --- | --- | --- | | language-adaptive modeling framework | 88.92 | Not recorded | 69eaf92b55a3c32af89f89fb | | language-adaptive modeling framework | 88.74 | Not recorded | 69eaf92b55a3c32af89f89fb | | language-adaptive modeling framework | 84.46 | Not recorded | 69eaf92b55a3c32af89f89fb | | language-adaptive modeling framework | 83.08 | Not recorded | 69eaf92b55a3c32af89f89fb | | language-adaptive modeling framework | 82.94 | Not recorded | 69eaf92b55a3c32af89f89fb | | language-adaptive modeling framework | 81.68 | Not recorded | 69eaf92b55a3c32af89f89fb | | Best of Sub 1 + Sub 4 | 81.2 | Not recorded | 69fc1cc35497b3b670c88781 | | Ensemble + threshold | 81.1 | Not recorded | 69fc1cc35497b3b670c88781 | | language-adaptive modeling framework | 80.61 | Not recorded | 69eaf92b55a3c32af89f89fb | | language-adaptive modeling framework | 80.32 | Not recorded | 69eaf92b55a3c32af89f89fb | | Best Gemma-12B per lang | 79.7 | Not recorded | 69fc1cc35497b3b670c88781 | | language-adaptive modeling framework | 79.58 | Not recorded | 69eaf92b55a3c32af89f89fb | | language-adaptive modeling framework | 79.03 | Not recorded | 69eaf92b55a3c32af89f89fb | | language-adaptive modeling framework | 78.23 | Not recorded | 69eaf92b55a3c32af89f89fb | | language-adaptive modeling framework | 77.44 | Not recorded | 69eaf92b55a3c32af89f89fb | | language-adaptive modeling framework | 76.32 | Not recorded | 69eaf92b55a3c32af89f89fb | | Hybrid (XLM-R, Qwen3) | 66.5 | Not recorded | 69fc1cc35497b3b670c88781 | ### language-adaptive modeling framework Macro-F1: 88.92. Result ID: 69eb0c9c47f34ce924d29c32. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Language":"Telugu (tel)"}. This record does not assert independent reproduction. ### language-adaptive modeling framework Macro-F1: 88.74. Result ID: 69eb0c9c47f34ce924d29c31. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Language":"Burmese (mya)"}. This record does not assert independent reproduction. ### language-adaptive modeling framework Macro-F1: 84.46. Result ID: 69eb0c9c47f34ce924d29c34. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Language":"Bengali (ben)"}. This record does not assert independent reproduction. ### language-adaptive modeling framework Macro-F1: 83.08. Result ID: 69eb0c9c47f34ce924d29c30. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Language":"Persian (fas)"}. This record does not assert independent reproduction. ### language-adaptive modeling framework Macro-F1: 82.94. Result ID: 69eb0c9c47f34ce924d29c35. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Language":"Arabic (arb)"}. This record does not assert independent reproduction. ### language-adaptive modeling framework Macro-F1: 81.68. Result ID: 69eb0c9c47f34ce924d29c33. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Language":"Hausa (hau)"}. This record does not assert independent reproduction. ### Best of Sub 1 + Sub 4 Macro-F1: 81.2. Result ID: 69fc62f947f34ce924df9ee5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Submission ID":"5"}. This record does not assert independent reproduction. ### Ensemble + threshold Macro-F1: 81.1. Result ID: 69fc62f947f34ce924df9ee4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Submission ID":"4"}. This record does not assert independent reproduction. ### language-adaptive modeling framework Macro-F1: 80.61. Result ID: 69eb0c9c47f34ce924d29c3b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Language":"Polish (pol)"}. This record does not assert independent reproduction. ### language-adaptive modeling framework Macro-F1: 80.32. Result ID: 69eb0c9c47f34ce924d29c36. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Language":"Hindi (hin)"}. This record does not assert independent reproduction. ### Best Gemma-12B per lang Macro-F1: 79.7. Result ID: 69fc62f947f34ce924df9ee2. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Submission ID":"1"}. This record does not assert independent reproduction. ### language-adaptive modeling framework Macro-F1: 79.58. Result ID: 69eb0c9c47f34ce924d29c39. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Language":"English (eng)"}. This record does not assert independent reproduction. ### language-adaptive modeling framework Macro-F1: 79.03. Result ID: 69eb0c9c47f34ce924d29c3a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Language":"Odia (ori)"}. This record does not assert independent reproduction. ### language-adaptive modeling framework Macro-F1: 78.23. Result ID: 69eb0c9c47f34ce924d29c38. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Language":"Swahili (swa)"}. This record does not assert independent reproduction. ### language-adaptive modeling framework Macro-F1: 77.44. Result ID: 69eb0c9c47f34ce924d29c37. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Language":"Amharic (amh)"}. This record does not assert independent reproduction. ### language-adaptive modeling framework Macro-F1: 76.32. Result ID: 69eb0c9c47f34ce924d29c3c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Language":"Spanish (spa)"}. This record does not assert independent reproduction. ### Hybrid (XLM-R, Qwen3) Macro-F1: 66.5. Result ID: 69fc62f947f34ce924df9ee3. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Submission ID":"2"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.