Planning on 8x8 Grid
89Validity RateL-ICL
Evaluation results
| Method | Links | ||
|---|---|---|---|
| L-ICLNumber of training examples (m)=60, Input representation (ASCII grid)=true2026.01 | 89 | 89 | |
| L-ICLNumber of training examples (m)=60, Input representation (ASCII grid)=false2026.01 | 73 | 63 | |
| Self-ConsistencyReasoning samples (k)=5, Input representation (ASCII grid)=true2026.01 | 59 | 45 | |
| ReActInference feedback=Oracle feedback, Input representation (ASCII grid)=true2026.01 | 55 | 45 | |
| Self-RefineReasoning samples (k)=5, Input representation (ASCII grid)=true2026.01 | 51 | 44 | |
| ReActPrompting strategy=ReAct (Prompt-Only), Input representation (ASCII grid)=true2026.01 | 48 | 41 | |
| L-ICLNumber of training examples (m)=0, Input representation (ASCII grid)=true2026.01 | 40 | 33 | |
| ToTPrompting strategy=ToT (Prompt-Only), Input representation (ASCII grid)=true2026.01 | 33 | 12 | |
| RAG-ICLPrompting strategy=RAG-ICL, Context size (characters)=20k, Input representation (ASCII grid)=true2026.01 | 21 | 9 | |
| RAG-ICLPrompting strategy=RAG-ICL, Context size (characters)=10k, Input representation (ASCII grid)=true2026.01 | 20 | 6 | |
| L-ICLNumber of training examples (m)=0, Input representation (ASCII grid)=false2026.01 | 19 | 12 | |
| Zero-ShotPrompting strategy=Zero-Shot, Input representation (ASCII grid)=true2026.01 | 16 | 0 |