# Self-Refining Video Sampling

arXiv: https://arxiv.org/abs/2601.18577 · HTML: https://mlsanity.com/paper/2601.18577/

## Key facts

- Cited by 15 papers in the map (data through 14 Oct 2026).
- Builds on 58 of its 117 references that are in the map.
- First posted on arXiv 26 Jan 2026.

## Builds on (most cited first, 30 of 58)

- [Denoising Diffusion Implicit Models](https://mlsanity.com/paper/2010.02502/) (2020)
- [Score-Based Generative Modeling through Stochastic Differential Equations](https://mlsanity.com/paper/2011.13456/) (2020)
- [Classifier-Free Diffusion Guidance](https://mlsanity.com/paper/2207.12598/) (2022)
- [Flow Matching for Generative Modeling](https://mlsanity.com/paper/2210.02747/) (2022)
- [Qwen2.5-VL Technical Report](https://mlsanity.com/paper/2502.13923/) (2025)
- [Generative Modeling by Estimating Gradients of the Data Distribution](https://mlsanity.com/paper/1907.05600/) (2019)
- [Self-Refine: Iterative Refinement with Self-Feedback](https://mlsanity.com/paper/2303.17651/) (2023)
- [Wan: Open and Advanced Large-Scale Video Generative Models](https://mlsanity.com/paper/2503.20314/) (2025)
- [CogVideoX: Text-to-Video Diffusion Models with An Expert Transformer](https://mlsanity.com/paper/2408.06072/) (2024)
- [SDEdit: Guided Image Synthesis and Editing with Stochastic Differential Equations](https://mlsanity.com/paper/2108.01073/) (2021)
- [VBench: Comprehensive Benchmark Suite for Video Generative Models](https://mlsanity.com/paper/2311.17982/) (2023)
- [HunyuanVideo: A Systematic Framework For Large Video Generative Models](https://mlsanity.com/paper/2412.03603/) (2024)
- [Training Deep Nets with Sublinear Memory Cost](https://mlsanity.com/paper/1604.06174/) (2016)
- [LTX-Video: Realtime Video Latent Diffusion](https://mlsanity.com/paper/2501.00103/) (2024)
- [UniPC: A Unified Predictor-Corrector Framework for Fast Sampling of Diffusion Models](https://mlsanity.com/paper/2302.04867/) (2023)
- [Pyramidal Flow Matching for Efficient Video Generative Modeling](https://mlsanity.com/paper/2410.05954/) (2024)
- [Improving Video Generation with Human Feedback](https://mlsanity.com/paper/2501.13918/) (2025)
- [MegaSaM: Accurate, Fast, and Robust Structure and Motion from Casual Dynamic Videos](https://mlsanity.com/paper/2412.04463/) (2024)
- [Video models are zero-shot learners and reasoners](https://mlsanity.com/paper/2509.20328/) (2025)
- [How Far is Video Generation from World Model: A Physical Law Perspective](https://mlsanity.com/paper/2411.02385/) (2024)
- [Motion-I2V: Consistent and Controllable Image-to-Video Generation with Explicit Motion Modeling](https://mlsanity.com/paper/2401.15977/) (2024)
- [World Simulation with Video Foundation Models for Physical AI](https://mlsanity.com/paper/2511.00062/) (2025)
- [PhysGen: Rigid-Body Physics-Grounded Image-to-Video Generation](https://mlsanity.com/paper/2409.18964/) (2024)
- [Cosmos-Reason1: From Physical Common Sense To Embodied Reasoning](https://mlsanity.com/paper/2503.15558/) (2025)
- [Gen2Act: Human Video Generation in Novel Scenarios enables Generalizable Robot Manipulation](https://mlsanity.com/paper/2409.16283/) (2024)
- [VideoPhy-2: A Challenging Action-Centric Physical Commonsense Evaluation in Video Generation](https://mlsanity.com/paper/2503.06800/) (2025)
- [Eliminating Oversaturation and Artifacts of High Guidance Scales in Diffusion Models](https://mlsanity.com/paper/2410.02416/) (2024)
- [Generalized Denoising Auto-Encoders as Generative Models](https://mlsanity.com/paper/1305.6663/) (2013)
- [VideoJAM: Joint Appearance-Motion Representations for Enhanced Motion Generation in Video Models](https://mlsanity.com/paper/2502.02492/) (2025)
- [FreeInit: Bridging Initialization Gap in Video Diffusion Models](https://mlsanity.com/paper/2312.07537/) (2023)

## Built on by (most cited first, 15 of 15)

- [Proprio: Latent Self-Scoring and Inference-Time Refinement for Physically Plausible Video Generation](https://mlsanity.com/paper/2605.28230/) (2026)
- [Human Cognition in Machines: A Unified Perspective of World Models](https://mlsanity.com/paper/2604.16592/) (2026)
- [VLMs are Good Teachers for Video Reasoning via Adaptive Test-Time Optimization](https://mlsanity.com/paper/2606.02564/) (2026)
- [CollabVR: Collaborative Video Reasoning with Vision-Language and Video Generation Models](https://mlsanity.com/paper/2605.08735/) (2026)
- [Steering Video Diffusion Transformers with Massive Activations](https://mlsanity.com/paper/2603.17825/) (2026)
- [Off-Manifold Refinement: Guiding Video Generators with a Frozen World Model](https://mlsanity.com/paper/2608.29904/) (2026)
- [On the Robustness of Distribution Support under Diffusion Guidance](https://mlsanity.com/paper/2605.07220/) (2026)
- [h-control: Training-Free Camera Control via Block-Conditional Gibbs Refinement](https://mlsanity.com/paper/2605.11871/) (2026)
- [SpecLoR: Spectral Lookahead Rectification for Motion-Coherent Text-to-Video Generation](https://mlsanity.com/paper/2606.11969/) (2026)
- [Test-Time Noise Guided Adaptation for Realistic Autoregressive Video Generation](https://mlsanity.com/paper/2607.15849/) (2026)
- [Reflection-aware Generative Novel View Synthesis](https://mlsanity.com/paper/2609.05382/) (2026)
- [Sampling headroom is not selection gain: a compute-value audit of test-time scaling for video world models](https://mlsanity.com/paper/2609.13257/) (2026)
- [Learning via Self-Consistency for Diffusion-based Video Reasoning](https://mlsanity.com/paper/2609.36826/) (2026)
- [HiPhy: Hierarchical Alignment for Physically-Plausible Multi-Principle Video Generation](https://mlsanity.com/paper/2610.02197/) (2026)
- [TrajCFG: Token-Wise Classifier-Free Guidance via Denoising Trajectory Feedback](https://mlsanity.com/paper/2610.14153/) (2026)

## Abstract (arXiv)

Modern video generators still struggle with complex physical dynamics, often falling short of physical realism. Existing approaches address this using external verifiers or additional training on augmented data, which is computationally expensive and still limited in capturing fine-grained motion. In this work, we present self-refining video sampling, a simple method that uses a pre-trained video generator trained on large-scale datasets as its own self-refiner. By interpreting the generator as a denoising autoencoder, we enable iterative inner-loop refinement at inference time without any external verifier or additional training. We further introduce an uncertainty-aware refinement strategy that selectively refines regions based on self-consistency, which prevents artifacts caused by over-refinement. Experiments on state-of-the-art video generators demonstrate significant improvements in motion coherence and physics alignment, achieving over 70% human preference compared to the default sampler and guidance-based sampler.