Research map / Large language models / AI and society
X-VoE: Measuring eXplanatory Violation of Expectation in Physical Events
Bo Dai, Linge Wang, Baoxiong Jia, Zeyu Zhang, Song-Chun Zhu, Chi Zhang, Yixin Zhu
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add residual connection to the dynamics module (ablation) | X-VoE model · X-VoE Collision and Blocking (holistic metric) | holistic metric | Improved |
| add residual connection to the dynamics module (ablation) | X-VoE model · X-VoE hypothetical setting (Collision, Blocking) | comparative metric | No clear effect |
| add occluded-object reasoning module (Transformers) on top of PLATO | PLATO · X-VoE explicative setting | comparative metric | Improved |
| use dynamics-based surprise s_dyn instead of image-based s_img (ablation) | X-VoE model · X-VoE Collision and Blocking (holistic metric) | holistic metric | Improved |
| use image-based surprise s_img instead of s_dyn (ablation) | X-VoE model · X-VoE Permanence and Continuity (holistic metric) | holistic metric | Improved |
| replace PLATO with the X-VoE model | PLATO · X-VoE Continuity (holistic metric) | holistic metric | No clear effect |
| increase training data size (ablation) | dynamics module · X-VoE predictive test videos | pixel loss | Improved |
| add follow-up information (lifting the wall) in explicative vs hypothetical setting (ablation) | X-VoE model · X-VoE Collision | comparative metric: gap close to 90% | Improved |
Closest papers by meaning
- Modeling Expectation Violation in Intuitive Physics with Coarse Probabilistic Object Representations (NeurIPS 2019 · similarity 0.89)
- Beyond Static Vision: Scene Dynamic Field Unlocks Intuitive Physics Understanding in Multi-modal Large Language Models (ICLR 2026 · similarity 0.85)
- Latent Intuitive Physics: Learning to Transfer Hidden Physics from A 3D Video (ICLR 2024 · similarity 0.84)
- VideoHallu: Evaluating and Mitigating Multi-modal Hallucinations on Synthetic Video Understanding (NeurIPS 2025 · similarity 0.84)
- QUANTIPHY: A Quantitative Benchmark Evaluating Physical Reasoning Abilities of Vision-Language Models (CVPR 2026 · similarity 0.83)
- PhysBench: Benchmarking and Enhancing Vision-Language Models for Physical World Understanding (ICLR 2025 · similarity 0.83)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
