Span-level Hallucination Detection on RAGTruth (test)
57.4F1 ScoreLettuceDetect-Qwen-2B
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| LettuceDetect-Qwen-2B2026.07 | 57.4 | 60.1 | 54.8 | 81.8 | |
| LFM-8B2026.07 | 55.7 | — | — | — | |
| LD-mmBERT2026.07 | 52.8 | — | — | — |
| Method | Links | ||||
|---|---|---|---|---|---|
| LettuceDetect-Qwen-2B2026.07 | 57.4 | 60.1 | 54.8 | 81.8 | |
| LFM-8B2026.07 | 55.7 | — | — | — | |
| LD-mmBERT2026.07 | 52.8 | — | — | — |