Toxicity Detection (Stage II) on TOXICTAGS (test)
64.66mF1STEMTOX
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| STEMTOXmodel=PALIGEMMA, task/shot=multi2025.08 | 64.66 | 72.4 | |
| LHLmodel=PALIGEMMA, task/shot=multi2025.08 | 61.54 | 69.2 | |
| STEMTOXmodel=LLAVA, task/shot=multi2025.08 | 57.99 | 66.9 | |
| LHLmodel=LLAVA, task/shot=multi2025.08 | 54.56 | 64.4 | |
| FS (TS)model=GPT, task/shot=4-shot2025.08 | 51.78 | 55.76 | |
| STEMTOXmodel=LLAVA, task/shot=uni2025.08 | 51.19 | 61.6 | |
| FS (TS)model=GPT, task/shot=2-shot2025.08 | 48.59 | 51.7 | |
| LHLmodel=PALIGEMMA, task/shot=uni2025.08 | 45.92 | 57.8 | |
| LHLmodel=LLAVA, task/shot=uni2025.08 | 44.69 | 58.7 | |
| STEMTOXmodel=PALIGEMMA, task/shot=uni2025.08 | 42.78 | 52.9 |