Knowledge Pair-wise Diversity on Empathetic Dialogue (test)
89.76PrecisionGTE
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| GTE2025.11 | 89.76 | 89.49 | 89.58 | 26.17 | 8.49 | 23.6 | |
| SBERT2025.11 | 89.32 | 89.32 | 89.32 | 28.73 | 10.88 | 26.41 | |
| Instructor2025.11 | 89.28 | 89.46 | 89.39 | 23.8 | 8.16 | 22.36 | |
| ChatGPT Inference2025.11 | 88.97 | 88.55 | 88.75 | 21.04 | 4.61 | 18.1 | |
| Contriever2025.11 | 88.93 | 88.62 | 88.77 | 23.64 | 7.36 | 21.25 | |
| DPR2025.11 | 86.81 | 86.67 | 86.72 | 14.39 | 2.94 | 13.53 | |
| MCTS-Driven Knowledge RetrievalConcept Bridging=false2025.11 | 86.53 | 86.26 | 86.41 | 13.29 | 2.76 | 11.83 | |
| MCTS-Driven Knowledge RetrievalReasoner=false2025.11 | 85.09 | 85.09 | 85.08 | 10.35 | 1.27 | 9.21 | |
| MCTS-Driven Knowledge Retrieval2025.11 | 84.71 | 84.72 | 84.7 | 9.64 | 1.05 | 8.66 |