ICCV 2025 accepted papers on Probabilistic and scientific modeling
30 of the 2,701 papers accepted at ICCV 2025 (1.1%) are on Probabilistic and scientific modeling, compared with 1.1% at ICCV 2023. They fall into 6 topics.
See ICCV 2025 in the interactive map
Submitting to ICCV? Checklist →
Working on something in this area? Describe your idea in scime atlas to see the closest papers by meaning and how crowded the spot has become.
Topics and papers
Molecules and proteins · 11 papers
- MolParser: End-to-end Visual Recognition of Molecule Structures in the Wild (27 citations)
- Future-Aware Interaction Network For Motion Forecasting (7 citations)
- PriorMotion: Generative Class-Agnostic Motion Prediction with Raster-Vector Motion Field Priors (7 citations)
- AMD: Adaptive Momentum and Decoupled Contrastive Learning Framework for Robust Long-Tail Trajectory Prediction (6 citations)
- Generative Active Learning for Long-tail Trajectory Prediction via Controllable Diffusion Model (4 citations)
- Towards Safer and Understandable Driver Intention Prediction (4 citations)
- NATRA: Noise-Agnostic Framework for Trajectory Prediction with Noisy Observations (3 citations)
- TOTP: Transferable Online Pedestrian Trajectory Prediction with Temporal-Adaptive Mamba Latent Diffusion (3 citations)
- From Gaze to Movement: Predicting Visual Attention for Autonomous Driving Human-Machine Interaction based on Programmatic Imitation Learning (2 citations)
- G2PDiffusion: Cross-Species Genotype-to-Phenotype Prediction via Evolutionary Diffusion (1 citation)
- Integrating Biological Knowledge for Robust Microscopy Image Profiling on De Novo Cell Lines
Variational and generative sampling · 8 papers
- Contrastive Flow Matching (31 citations)
- The Curse of Conditions: Analyzing and Improving Optimal Transport for Conditional Flow-Based Generation (9 citations)
- Riemannian-Geometric Fingerprints of Generative Models (7 citations)
- Taming Flow Matching with Unbalanced Optimal Transport into Fast Pansharpening (6 citations)
- Improving Rectified Flow with Boundary Conditions (3 citations)
- Flow Stochastic Segmentation Networks (2 citations)
- SCFlow: Implicitly Learning Style and Content Disentanglement with Flow Models (1 citation)
- Stochastic Interpolants for Revealing Stylistic Flows across the History of Art
Causal discovery · 3 papers
- CWNet: Causal Wavelet Network for Low-Light Image Enhancement (43 citations)
- Learning Counterfactually Decoupled Attention for Open-World Model Attribution (3 citations)
- C2MIL: Synchronizing Semantic and Topological Causalities in Multiple Instance Learning for Robust and Interpretable Survival Analysis (1 citation)
Dynamical systems · 3 papers
Differential equations and PDEs · 3 papers
- HazeFlow: Revisit Haze Physical Model as ODE and Non-Homogeneous Haze Generation for Real-World Dehazing (13 citations)
- ResQ: A Novel Framework to Implement Residual Neural Networks on Analog Rydberg Atom Quantum Computers (2 citations)
- FuXi-RTM: A Physics-Guided Prediction Framework with Radiative Transfer Modeling
