D-JEPA: Design-Recoverable JEPA Representation with Swappable Physics Decoders
cs.LG, cs.AI
Submitted: 2026-09-27
Updated: 2026-09-27
Terminology
Sources
- GeoTransolver: Learning Physics on Irregular Domains Using Multi-scale Geometry Aware Physics Attention Transformer
- LeJEPA: Provable and Scalable Self-Supervised Learning Without the Heuristics
- Revisiting Feature Prediction for Learning Visual Representations from Video
- AeroJEPA: Learning Semantic Latent Representations for Scalable 3D Aerodynamic Field Modeling
- DoMINO: A Decomposable Multi-scale Iterative Neural Operator for Modeling Large Scale Engineering Simulations
- AutoHood3D: A Multi-Modal Benchmark for Automotive Hood Design and Fluid-Structure Interaction
Related papers
- Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
- AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
- Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
- Convergence issues in Relational Concept Analysis based on AOC-posets
- Beliefs Beyond Posteriors: Local-Consistency Optimisation for Bayesian Neural Networks
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks