Clarification Seeking on WebShop Clarification (test)
0.245Success RateReAct+UE
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| ReAct+UEModel=GPT-5.12026.06 | 0.245 | 0.235 | 0.085 | 0.125 | 0.404 | |
| Uncertainty DecompositionModel=GLM-4.72026.06 | 0.19 | 0.694 | 0.829 | 0.756 | 0.722 | |
| UAMModel=GPT-5.12026.06 | 0.189 | 0.238 | 0.102 | 0.143 | 0.368 | |
| Uncertainty DecompositionModel=GPT-5.12026.06 | 0.136 | 0.267 | 0.105 | 0.151 | 0.444 | |
| UAMModel=GLM-4.72026.06 | 0.114 | 0.524 | 0.268 | 0.355 | 0.494 | |
| ReAct+UEModel=GLM-4.72026.06 | 0.112 | 0.625 | 0.3 | 0.405 | 0.551 | |
| Uncertainty DecompositionModel=DeepSeek-v3.2-exp2026.06 | 0.08 | 0.547 | 0.7 | 0.614 | 0.56 | |
| ReAct+UEModel=DeepSeek-v3.2-exp2026.06 | 0.07 | 0.139 | 0.1 | 0.116 | 0.24 | |
| Uncertainty DecompositionModel=Qwen3.5-35B2026.06 | 0.064 | 0.5 | 0.511 | 0.505 | 0.5 | |
| ReAct+UEModel=Qwen3.5-35B2026.06 | 0.051 | 0.097 | 0.06 | 0.074 | 0.242 | |
| Uncertainty DecompositionModel=GPT-OSS-120B2026.06 | 0.044 | 0.245 | 0.255 | 0.25 | 0.209 | |
| UAMModel=DeepSeek-v3.2-exp2026.06 | 0.04 | 0.173 | 0.18 | 0.176 | 0.16 | |
| UAMModel=GPT-OSS-120B2026.06 | 0.033 | 0.217 | 0.222 | 0.22 | 0.211 | |
| ReAct+UEModel=GPT-OSS-120B2026.06 | 0.023 | 0.184 | 0.149 | 0.165 | 0.174 | |
| UAMModel=Qwen3.5-35B2026.06 | 0.02 | 0.178 | 0.16 | 0.168 | 0.21 |