Visual Planning on COIN
33.99Success Rate (SR)ViterbiPlanNet
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| ViterbiPlanNetParams (M)=6.672026.03 | 33.99 | 50.87 | 83.88 | |
| PlanLLMParams (M)=386.432026.03 | 33.44 | 51.05 | 84.66 | |
| SCHEMAParams (M)=6.282026.03 | 32.89 | 50.84 | 83.98 | |
| ViterbiPlanNetParams (M)=6.872026.03 | 23.92 | 45.63 | 82.56 | |
| PlanLLMParams (M)=386.432026.03 | 23.19 | 45.7 | 83.44 | |
| PDPPParams (M)=42.902026.03 | 22.37 | 44.6 | 83 | |
| SCHEMAParams (M)=6.282026.03 | 22.33 | 45.21 | 82.93 | |
| Gemini 2.5 ProParams (M)=>100,0002026.03 | 17.02 | 38.87 | 78.73 | |
| PDPPParams (M)=42.902026.03 | 15.21 | 41.01 | 81.64 | |
| KEPPParams (M)=44.662026.03 | 15.2 | 33.39 | 67.79 | |
| Qwen2.5-32BParams (M)=32,0002026.03 | 14.97 | 36.34 | 78.74 | |
| Qwen3-30B + PKGParams (M)=30,0002026.03 | 14.63 | 36.53 | 78.11 | |
| Qwen3-30BParams (M)=30,0002026.03 | 14.52 | 36.56 | 78.07 | |
| KEPPParams (M)=44.662026.03 | 13.85 | 28.4 | 62.54 | |
| PKG beam searchParams (M)=42.902026.03 | 13.32 | 37.42 | 78.93 | |
| Gemini 2.5 ProParams (M)=>100,0002026.03 | 8.1 | 31.9 | 71.7 | |
| PKG beam searchParams (M)=42.902026.03 | 5.14 | 31.29 | 74.26 | |
| Qwen2.5-32BParams (M)=32,0002026.03 | 4.98 | 27.45 | 71.64 | |
| Qwen3-30B + PKGParams (M)=30,0002026.03 | 4.78 | 29 | 71.04 | |
| Qwen3-30BParams (M)=30,0002026.03 | 4.64 | 28.85 | 70.45 | |
| Qwen2.5-VL-32BParams (M)=32,0002026.03 | 3.65 | 17.51 | 52.1 | |
| Qwen2.5-VL-32BParams (M)=32,0002026.03 | 1.87 | 17.05 | 55.66 |