Question Answering on 258 banking QA examples (test)
6.21JudgeLM ScoreFinRAG-12B
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| FinRAG-12B2026.05 | 6.21 | 73.1 | 93.6 | 12 | |
| GPT-4.1access=API2026.05 | 5.72 | 70.8 | 90 | 20.2 | |
| Gemma 3 12Bfine-tuning=none2026.05 | 5.7 | 80.2 | 96.4 | 4.3 |
| Method | Links | ||||
|---|---|---|---|---|---|
| FinRAG-12B2026.05 | 6.21 | 73.1 | 93.6 | 12 | |
| GPT-4.1access=API2026.05 | 5.72 | 70.8 | 90 | 20.2 | |
| Gemma 3 12Bfine-tuning=none2026.05 | 5.7 | 80.2 | 96.4 | 4.3 |