Embodied AI Planning on TDW-GAME
100Success RateBaseline
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| BaselineModel Scale=GPT-5, Execution Strategy=Baseline2026.04 | 100 | 7.88 | 2.3 | 14.3 | |
| AgenticCache (Ours+)Model Scale=GPT-5, Execution Strategy=AgenticCache (Ours+)2026.04 | 100 | 1.11 | 728 | 4.8 | |
| AgenticCache (Ours+)Model Scale=GPT-5-mini, Execution Strategy=AgenticCache (Ours+)2026.04 | 100 | 1.13 | 840 | 1 | |
| AgenticCache (Ours+)Model Scale=GPT-5-nano, Execution Strategy=AgenticCache (Ours+)2026.04 | 100 | 1.08 | 1.2 | 0.4 | |
| SpeculativeModel Scale=GPT-5-mini, Execution Strategy=Speculative2026.04 | 33.33 | 4.61 | 8 | 4.7 | |
| BaselineModel Scale=GPT-5-mini, Execution Strategy=Baseline2026.04 | 22.22 | 11.06 | 3.9 | 4.5 | |
| ParallelModel Scale=GPT-5-mini, Execution Strategy=Parallel2026.04 | 22.22 | 6.08 | 6 | 7.5 | |
| SpeculativeModel Scale=GPT-5, Execution Strategy=Speculative2026.04 | 11.11 | 6.13 | 9.9 | 28.5 | |
| ParallelModel Scale=GPT-5, Execution Strategy=Parallel2026.04 | 0 | 15.83 | 10.8 | 84.2 | |
| BaselineModel Scale=GPT-5-nano, Execution Strategy=Baseline2026.04 | 0 | 10.07 | 7.8 | 2.4 | |
| ParallelModel Scale=GPT-5-nano, Execution Strategy=Parallel2026.04 | 0 | 6.48 | 12.4 | 3.9 |