sampling efficiency
A measure of how effectively an AI model can generate high-quality samples from a distribution with a limited number of samples drawn. This is particularly important in contexts such as generative modeling and reinforcement learning.
- Adjoint Schrödinger Bridge Sampler
- Adjoint Schrödinger Bridge Sampler
- Beyond Scores: Proximal Diffusion Models
- Convex Potential Mirror Langevin Algorithm for Efficient Sampling of Energy-Based Models
- Cross-fluctuation phase transitions reveal sampling dynamics in diffusion models
- Discovering Important Experts for Mixture-of-Experts Models Pruning Through a Theoretical Perspective
- Does Reinforcement Learning Really Incentivize Reasoning Capacity in LLMs Beyond the Base Model?
- Does Reinforcement Learning Really Incentivize Reasoning Capacity in LLMs Beyond the Base Model?
- Exploring the Design Space of Diffusion Bridge Models
- FAST: Foreground‑aware Diffusion with Accelerated Sampling Trajectory for Segmentation‑oriented Anomaly Synthesis
- FSEO: Few-Shot Evolutionary Optimization via Meta-Learning for Expensive Multi-Objective Optimization
- Flow-GRPO: Training Flow Matching Models via Online RL
- Generative diffusion for perceptron problems: statistical physics analysis and efficient algorithms
- Learnable Sampler Distillation for Discrete Diffusion Models
- MDNS: Masked Diffusion Neural Sampler via Stochastic Optimal Control
- MRO: Enhancing Reasoning in Diffusion Language Models via Multi-Reward Optimization
- Masked Diffusion Models as Energy Minimization
- Neural Stochastic Flows: Solver-Free Modelling and Inference for SDE Solutions
- NeuralPLexer3: Accurate Biomolecular Complex Structure Prediction with Flow Models
- Path Gradients after Flow Matching
- SCoT: Unifying Consistency Models and Rectified Flows via Straight-Consistent Trajectories
- Straight-Line Diffusion Model for Efficient 3D Molecular Generation
- TPP-SD: Accelerating Transformer Point Process Sampling with Speculative Decoding