Comparative Sentiment Classification on SUDO (test)
60.51Micro PrecisionXCom
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| XCom2026.03 | 60.51 | 56.4 | 58.38 | 61.92 | 56.11 | 58.56 | |
| Finetuned-T5Model Category=Transformer-based baselines, Evaluation Protocol=Finetuned2026.03 | 55.88 | 57.72 | 56.79 | 55.24 | 56.53 | 55.86 | |
| FastText+SVMModel Category=Feature-based baselines2026.03 | 52.57 | 49.15 | 50.08 | 52.83 | 49.76 | 51.05 | |
| FastText+XGBoostModel Category=Feature-based baselines2026.03 | 52.03 | 42.28 | 46.65 | 53.65 | 41.38 | 46.22 | |
| Finetuned-BARTModel Category=Transformer-based baselines, Evaluation Protocol=Finetuned2026.03 | 49.7 | 61.49 | 54.97 | 50.16 | 62.74 | 55.72 | |
| ChatGPT-5.1Model Category=General-purpose LLM baselines2026.03 | 47.28 | 50.33 | 48.76 | 47.61 | 54.33 | 47.24 | |
| Gemini-2.5-FlashModel Category=General-purpose LLM baselines2026.03 | 46.06 | 48.96 | 47.47 | 48.97 | 53.39 | 44.94 | |
| Llama-3.2-8B-InstructModel Category=General-purpose LLM baselines2026.03 | 24.86 | 33.9 | 28.69 | 25.62 | 34.69 | 25.61 |