Open X-Embodiment
1M+ real-robot trajectories from 20+ institutions.
Optimism Bias World Model Benchmark — Training Data Training data for VLM-as-judge models evaluating world model predictions. Structure ├── part1…
Optimism Bias World Model Benchmark — Training Data Training data for VLM-as-judge models evaluating world model predictions. Structure ├── part1 physconscore/ Physical Consistency Scoring ├── part2 action following/ Action Following Evaluation └── part3 optimism bias/ Optimism Bias Detection Part 1: Physical Consistency Score (PhysConsScore) Task: Binary classification (Good/Bad) on 16 physics indicators Model: Qwen3-VL-8B-Instruct (LoRA… See the full description on the dataset page: train.
Source: Hugging Face Hub (TANDREWYANG/optimism_train). Metadata imported from the dataset’s Hub tags.