Metaphor detection on MOH-X
89.7F1 ScoreLAG
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| LAGLLM=Claude 3.5 Sonnet2025.04 | 89.7 | 87.3 | — | — | |
| Gemini 2.5Training Status=Fine-tuned on VUA-182026.04 | 84.9 | 84.7 | 80.9 | 89.3 | |
| Gemini 2.5Evaluation protocol=zero-shot, Evaluation target=target words2026.04 | 84.9 | — | — | — | |
| MetaPRO2025.04 | 84 | 81 | — | — | |
| TSI CMT*2025.04 | 82.5 | 82.9 | — | — | |
| Claude Sonnet 4.0Training Status=Fine-tuned on VUA-182026.04 | 82.2 | 80.6 | 73.9 | 92.6 | |
| Claude Sonnet 4.0Evaluation protocol=zero-shot, Evaluation target=target words2026.04 | 82.2 | — | — | — | |
| ConceptFrameMetTraining Status=Fine-tuned on VUA-182026.04 | 81.4 | 80.3 | 74.6 | 89.9 | |
| ConceptFrameMetEvaluation protocol=fine-tuned, Training dataset=VUA-18 (train), Evaluation target=target words2026.04 | 81.4 | — | — | — | |
| MelBertTraining Status=Fine-tuned on VUA-182026.04 | 80.6 | 80 | 75.3 | 87.1 | |
| MelBertEvaluation protocol=fine-tuned, Training dataset=VUA-18 (train), Evaluation target=target words2026.04 | 80.6 | — | — | — | |
| FrameBertTraining Status=Fine-tuned on VUA-182026.04 | 78 | 76.9 | 72 | 85.9 | |
| FrameBertEvaluation protocol=fine-tuned, Training dataset=VUA-18 (train), Evaluation target=target words2026.04 | 78 | — | — | — | |
| Random baselineTraining Status=Fine-tuned on VUA-182026.04 | 48.6 | 48.4 | 46.8 | 50.5 | |
| Random baseline2026.04 | 48.6 | — | — | — |