Tool Selection on X ToolE out-of-domain (test)
89.87Recall@1L_PORTS
Evaluation Results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| L_PORTSEncoder=BGE, Best LLM=LLAMA3-GROQ-8B-Tool-Use2026.07 | 89.87 | 92.2 | 94.04 | 89.87 | 91.1 | 92.35 | 14.18 | 17.11 | |
| L_PORTSEncoder=RoBERTa, Best LLM=LLAMA3-8B2026.07 | 74.6 | 83.9 | 86.8 | 74.6 | 81.23 | 83.55 | 61.24 | 59.79 | |
| L_replugEncoder=BGE, Best LLM=LLAMA3-GROQ-8B-Tool-Use2026.07 | 74.6 | 86.24 | 90.3 | 74.6 | 83.98 | 85.09 | 5.85 | 7.22 | |
| L_replugEncoder=RoBERTa, Best LLM=LLAMA3-GROQ-8B-Tool-Use2026.07 | 58.51 | 68.54 | 74.1 | 58.51 | 67.62 | 69.78 | 46.53 | 45.3 |