A Benchmark for Modeling Violation-of-Expectation in Physical Reasoning Across Event Categories
This work provides a benchmark for physical reasoning in AI, addressing a domain-specific need for better interpretability and evaluation in cognitive models.
The authors tackled the problem of evaluating physical reasoning in AI by creating a large-scale synthetic 3D Violation-of-Expectation dataset with ground-truth heuristic labels, and they proposed OFPR-Net, which outperformed baselines and demonstrated flexibility in learning alternate physical realities.
Recent work in computer vision and cognitive reasoning has given rise to an increasing adoption of the Violation-of-Expectation (VoE) paradigm in synthetic datasets. Inspired by infant psychology, researchers are now evaluating a model's ability to label scenes as either expected or surprising with knowledge of only expected scenes. However, existing VoE-based 3D datasets in physical reasoning provide mainly vision data with little to no heuristics or inductive biases. Cognitive models of physical reasoning reveal infants create high-level abstract representations of objects and interactions. Capitalizing on this knowledge, we established a benchmark to study physical reasoning by curating a novel large-scale synthetic 3D VoE dataset armed with ground-truth heuristic labels of causally relevant features and rules. To validate our dataset in five event categories of physical reasoning, we benchmarked and analyzed human performance. We also proposed the Object File Physical Reasoning Network (OFPR-Net) which exploits the dataset's novel heuristics to outperform our baseline and ablation models. The OFPR-Net is also flexible in learning an alternate physical reality, showcasing its ability to learn universal causal relationships in physical reasoning to create systems with better interpretability.