Tool Selection on API-Bank (test)
59.12Recall@1L_PORTS
Evaluation Results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| L_PORTSEncoder=BGE, Best LLM=LLAMA3-GROQ-8B-Tool-Use2026.07 | 59.12 | 76.8 | 81.5 | 59.12 | 75.4 | 76.1 | 14.94 | 14.21 | |
| L_replugEncoder=BGE, Best LLM=LLAMA3-GROQ-8B-Tool-Use2026.07 | 56.29 | 75 | 80 | 56.29 | 70.6 | 73.32 | 12.9 | 10.73 | |
| L_PORTSEncoder=RoBERTa, Best LLM=LLAMA3-GROQ-8B-Tool-Use2026.07 | 49.84 | 64.35 | 70.8 | 49.84 | 62.27 | 65.32 | 57.1 | 54.14 | |
| L_replugEncoder=RoBERTa, Best LLM=LLAMA3-GROQ-8B-Tool-Use2026.07 | 45.32 | 61.94 | 68.23 | 45.32 | 59.95 | 62.44 | 53.93 | 50.9 |