Tool selection on CD Cross-domain
97.33AccuracyMTA
Evaluation Results
| Method | Links | |
|---|---|---|
| MTABackbone=Baichuan2-Chat (7B)2026.01 | 97.33 | |
| MTABackbone=Qwen1.5-Chat (7B)2026.01 | 97.16 | |
| FTBackbone=Baichuan2-Chat (7B)2026.01 | 96.89 | |
| FTBackbone=Qwen1.5-Chat (7B)2026.01 | 96.39 | |
| MTABackbone=ChatGLM3 (6B)2026.01 | 93.94 | |
| FTBackbone=Chinese Alpaca-2 (7B)2026.01 | 92.57 | |
| MTABackbone=Chinese Alpaca-2 (7B)2026.01 | 91.46 | |
| FTBackbone=ChatGLM3 (6B)2026.01 | 89.13 | |
| ReActBackbone=Qwen1.5-Chat (7B)2026.01 | 57.55 | |
| ReActBackbone=ChatGLM3 (6B)2026.01 | 51.15 | |
| ReActBackbone=Baichuan2-Chat (7B)2026.01 | 46.2 | |
| CoTBackbone=Qwen1.5-Chat (7B)2026.01 | 44.34 | |
| CoTBackbone=ChatGLM3 (6B)2026.01 | 42.29 | |
| BaseBackbone=Qwen1.5-Chat (7B)2026.01 | 40.82 | |
| BaseBackbone=ChatGLM3 (6B)2026.01 | 40.57 | |
| BaseBackbone=Baichuan2-Chat (7B)2026.01 | 39.17 | |
| CoTBackbone=Baichuan2-Chat (7B)2026.01 | 34.48 | |
| BaseBackbone=Chinese Alpaca-2 (7B)2026.01 | 29.86 | |
| CoTBackbone=Chinese Alpaca-2 (7B)2026.01 | 26.41 | |
| ReActBackbone=Chinese Alpaca-2 (7B)2026.01 | 25.82 |