Hallucination Detection on FinQA retrieval-equalized (test)
2.1P95 Latency (s)HHEM
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| HHEM2026.04 | 2.1 | 0.001 | 76.3 | |
| FINGROUNDModel Size=8B, Hardware=A1002026.04 | 3.8 | 0.003 | 91.4 | |
| FINGROUNDModel Size=8B, Hardware=A10G2026.04 | 5.9 | 0.002 | 91.4 | |
| GPT-4oModel Role=teacher2026.04 | 8.2 | 0.047 | 95 | |
| SelfCheckGPT2026.04 | 12.4 | 0.062 | 72.8 |