ProcTHOR
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Visually-Grounded Active View Selection | AVS-ProcTHOR (val) | Existence Score93.02 | 11 | |
| Object Search | ProcTHOR 150 distinct maps (test) | Average Navigation Cost215.59 | 9 | |
| Multi-Robot Planning | ProcTHOR Large | Average Cost429.83 | 9 | |
| Multi-Robot Planning | ProcTHOR Medium | Average Cost269.45 | 9 | |
| Multi-Robot Planning | ProcTHOR Small | Average Cost130.86 | 9 | |
| Action Prediction | ProcTHOR single-object (OOD Systematic) | Accuracy75 | 9 | |
| Action Prediction | ProcTHOR single-object (OOD Compositional) | Accuracy95 | 9 | |
| Action Prediction | ProcTHOR single-object (IID) | Accuracy97 | 9 | |
| Action Prediction | ProcTHOR multi-object | IID Accuracy94 | 8 | |
| Pairwise Scene Generation Evaluation | Procthor-10K (Easy) | Score (SceneCritic) Method A75.6 | 6 | |
| Pairwise evaluator agreement with human judgment | Procthor-10K Complex 1.0 (test) | Method A SceneCritic Score79.5 | 6 | |
| Interactive Navigation | ProcTHOR-10k 7-10 rooms (test) | SR100 | 6 | |
| Interactive Navigation | ProcTHOR-10k 4-6 rooms (test) | Success Rate100 | 6 | |
| Interactive Navigation | ProcTHOR-10k 1-3 rooms (test) | Success Rate (SR)100 | 6 | |
| Task Planning | PROCTHOR Any-of-Three | Average Cost22.03 | 5 | |
| Task Planning | PROCTHOR Breakfast+Coffee | Average Cost207.17 | 5 | |
| Task Planning | PROCTHOR Coffee | Avg Cost112.76 | 5 | |
| Task Planning | PROCTHOR Breakfast | Average Cost82.59 | 5 | |
| Task Planning | PROCTHOR Deliver 3-Object | Avg. Cost94.58 | 5 | |
| Navigation | ProcTHOR | Success Rate58 | 4 |
Showing 20 of 20 rows