AI Paper Atlas

Research map / Probabilistic and scientific modeling / Dynamical systems

On the Learning Mechanisms in Physical Reasoning

Shiqian Li, Kewen Wu, Chi Zhang, Yixin Zhu

NeurIPS 2022 · Probabilistic and scientific modeling › Dynamical systems · approach: Latent dynamics models · 0 citations (OpenAlex)

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 changedWhereMeasuredEffect
Use ViT to classify outcomes directly from initial scenes instead of predicting object dynamics with RPINPHYRE-B physical reasoning · PHYRE-B cross-templateAUCCESS: 50.86 -> 56.31Improved
Use ViT direct classification instead of RPIN dynamics predictionPHYRE-B physical reasoning · PHYRE-B within-templateAUCCESS: 85.49 -> 84.16$\pm$0.30Mixed
Supply eight ground-truth dynamics frames instead of intuition-only input (ablation)TimeSformer on PHYRE-B · PHYRE-B within-template and cross-templateAUCCESS: +23.11 within; +48.16 crossImproved
Replace ground-truth simulator dynamics with PredRNN-predicted frames using parallel optimization (ablation)TimeSformer on PHYRE-B · PHYRE-B within-template and cross-templateAUCCESS: Simulator 4 frames: 86.75, 73.81; PredRNN parallel: 75.22, 46.42Hurt
Train PredRNN serially and freeze it before training TimeSformer, instead of parallel optimization (ablation)PredRNN + TimeSformer on PHYRE-B · PHYRE-B within-template and cross-templateAUCCESS: 75.22 -> 64.90 within; 46.42 -> 44.33 crossHurt
Use predicted future frames rather than a single initial frame with TimeSformer (ablation)TimeSformer on PHYRE-B · PHYRE-B within-templateAUCCESS: 73.53 -> 75.22Improved
Use predicted future frames rather than a single initial frame with TimeSformer (ablation)TimeSformer on PHYRE-B · PHYRE-B cross-templateAUCCESS: 46.02 -> 46.42Mixed
Use Swin instead of ViT for direct image classification (ablation)LFI on PHYRE-B · PHYRE-B within-templateAUCCESS: 84.16$\pm$0.30 -> 84.71$\pm$0.33Improved
Use Swin instead of ViT for direct image classification (ablation)LFI on PHYRE-B · PHYRE-B cross-templateAUCCESS: 56.31$\pm$1.95 -> 54.92$\pm$2.30Hurt
Use BEiT instead of ViT for direct image classification (ablation)LFI on PHYRE-B · PHYRE-B within-templateAUCCESS: 84.16$\pm$0.30 -> 83.59$\pm$0.09Hurt
Use BEiT instead of ViT for direct image classification (ablation)LFI on PHYRE-B · PHYRE-B cross-templateAUCCESS: 56.31$\pm$1.95 -> 54.07$\pm$1.88Hurt

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy