Emotional Support Dialogue on ExTES
74.6SRw. UKA
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| w. UKABackbone=Seed-1.6-36B2026.05 | 74.6 | 6.48 | |
| w. MetaMindBackbone=Seed-1.6-36B2026.05 | 73.9 | 6.01 | |
| w. PRINCIPLESBackbone=Seed-1.6-36B2026.05 | 73.2 | 6.54 | |
| Prompting BaselineBackbone=Seed-1.6-36B2026.05 | 68.8 | 6.34 | |
| w. UKABackbone=Qwen3-32B2026.05 | 63.6 | 6.93 | |
| w. UKABackbone=Qwen3-235B-A22B2026.05 | 62 | 7.1 | |
| w. PRINCIPLESBackbone=Qwen3-32B2026.05 | 60.1 | 7.17 | |
| w. PRINCIPLESBackbone=Qwen3-235B-A22B2026.05 | 59.1 | 7.12 | |
| w. UKABackbone=GPT-OSS-120B2026.05 | 56.4 | 7.01 | |
| w. MetaMindBackbone=Qwen3-235B-A22B2026.05 | 55.5 | 7.19 | |
| w. MetaMindBackbone=GPT-OSS-120B2026.05 | 55.1 | 6.75 | |
| w. MetaMindBackbone=Qwen3-32B2026.05 | 53.3 | 7.28 | |
| Prompting BaselineBackbone=Qwen3-235B-A22B2026.05 | 52.9 | 7.21 | |
| Prompting BaselineBackbone=Qwen3-32B2026.05 | 51 | 7.32 | |
| w. PRINCIPLESBackbone=GPT-OSS-120B2026.05 | 48 | 7.23 | |
| Prompting BaselineBackbone=GPT-OSS-120B2026.05 | 35.7 | 7.48 |