Tool Retrieval on ToolBench Vague queries G1 tier
49.4NDCG@5CoHyDE (r = 3)
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| CoHyDE (r = 3)Expansion Factor (r)=32026.05 | 49.4 | 55.2 | |
| BGE (trained S1a)Backbone=BGE, Training Stage=S1a2026.05 | 44.7 | 49.8 | |
| HyDE (vanilla LLM + trained BGE S1a)Backbone=BGE (trained S1a), Query Expansion Strategy=HyDE, Generator=Qwen3.5-4B2026.05 | 41.2 | 47.4 | |
| HyDE (vanilla LLM & BGE)Backbone=BGE (vanilla), Query Expansion Strategy=HyDE, Generator=Qwen3.5-4B2026.05 | 37.4 | 43.4 | |
| text-embedding-3-largeBackbone=text-embedding-3-large2026.05 | 34.8 | 40.5 | |
| BGE (vanilla)Backbone=BGE, Training Stage=vanilla2026.05 | 33.2 | 37.3 | |
| Query expansion (LLM + BGE)Backbone=BGE (vanilla), Query Expansion Strategy=Query-side paraphrasing2026.05 | 32.6 | 38.2 | |
| BM252026.05 | 17.8 | 19.6 |