IR Compliance Evaluation on IR Compliance GFT
0Self Violation RateGem.-2-Pro
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Gem.-2-ProRole=Seller2026.04 | 0 | 0 | |
| o3Role=Buyer2026.04 | 5 | 20 | |
| o3Role=Seller2026.04 | 5 | 30 | |
| Gem.-2-FlashRole=Seller2026.04 | 10 | 10 | |
| Gem.-2-ProRole=Buyer2026.04 | 10 | 20 | |
| GPT-4oRole=Seller2026.04 | 10 | 20 | |
| Gem.-2-FlashRole=Buyer2026.04 | 20 | 25 | |
| GPT-4oRole=Buyer2026.04 | 25 | 15 | |
| DeepSeek-V3Role=Buyer2026.04 | 45 | 15 | |
| DeepSeek-V3Role=Seller2026.04 | 70 | 45 |