Occupancy Modeling via Drifting: One-Step Generation Enables Direct Policy Extraction
cs.LG
Submitted: 2026-10-14
Updated: 2026-10-14
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Sources
- Proto Successor Measure: Representing the Behavior Space of an RL Agent
- TD-JEPA: Latent-predictive Representations for Zero-Shot Reinforcement Learning
- Learning Successor States and Goal-Dependent Values: A Mathematical Viewpoint
- Exploration by Random Network Distillation
- Generative Modeling via Drifting
- A Unified Framework for Zero-Shot Reinforcement Learning
- Diversity is All You Need: Learning Skills without a Reward Function
- C-Learning: Learning to Achieve Goals via Recursive Classification
- Contrastive Learning as Goal-Conditioned Reinforcement Learning
- Temporal Difference Flows
- Compositional Planning with Jumpy World Models
- One Step Diffusion via Shortcut Models
- Drift-Based Policy Optimization: Native One-Step Policy Learning for Online Robot Control
- Mean Flows for One-step Generative Modeling
- Dream to Control: Learning Behaviors by Latent Imagination
- Learning Continuous Control Policies by Stochastic Value Gradients
- Drift Q-Learning
- Generative Temporal Difference Learning for Infinite-Horizon Prediction
- Drifting Field Policy: A One-Step Generative Policy via Wasserstein Gradient Flow
- Offline Reinforcement Learning with Implicit Q-Learning
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: The Predictive Cost of ELBO Binding
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks