# APLA: Additional Perturbation for Latent Noise with Adversarial Training Enables Consistency

arXiv: https://arxiv.org/abs/2308.12605 · HTML: https://mlsanity.com/paper/2308.12605/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- Placed in Video generation, one of 14 directions in Diffusion models.
- Builds on 39 of its 49 references that are in the map.
- First posted on arXiv 24 Aug 2023.

## Placed in

- Diffusion models / [Video generation](https://mlsanity.com/diffusion/video-generation/)

## Builds on (most cited first, 30 of 39)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [The Unreasonable Effectiveness of Deep Features as a Perceptual Metric](https://mlsanity.com/paper/1801.03924/) (2018)
- [Denoising Diffusion Implicit Models](https://mlsanity.com/paper/2010.02502/) (2020)
- [BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models](https://mlsanity.com/paper/2301.12597/) (2023)
- [Classifier-Free Diffusion Guidance](https://mlsanity.com/paper/2207.12598/) (2022)
- [Adding Conditional Control to Text-to-Image Diffusion Models](https://mlsanity.com/paper/2302.05543/) (2023)
- [Zero-Shot Text-to-Image Generation](https://mlsanity.com/paper/2102.12092/) (2021)
- [GLIDE: Towards Photorealistic Image Generation and Editing with Text-Guided Diffusion Models](https://mlsanity.com/paper/2112.10741/) (2021)
- [Taming Transformers for High-Resolution Image Synthesis](https://mlsanity.com/paper/2012.09841/) (2020)
- [Imagen Video: High Definition Video Generation with Diffusion Models](https://mlsanity.com/paper/2210.02303/) (2022)
- [ViViT: A Video Vision Transformer](https://mlsanity.com/paper/2103.15691/) (2021)
- [RePaint: Inpainting using Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2201.09865/) (2022)
- [CogVideo: Large-scale Pretraining for Text-to-Video Generation via Transformers](https://mlsanity.com/paper/2205.15868/) (2022)
- [The 2017 DAVIS Challenge on Video Object Segmentation](https://mlsanity.com/paper/1704.00675/) (2017)
- [VideoGPT: Video Generation using VQ-VAE and Transformers](https://mlsanity.com/paper/2104.10157/) (2021)
- [Generating Wikipedia by Summarizing Long Sequences](https://mlsanity.com/paper/1801.10198/) (2018)
- [Long Video Generation with Time-Agnostic VQGAN and Time-Sensitive Transformer](https://mlsanity.com/paper/2204.03638/) (2022)
- [Semi-supervised Sequence Learning](https://mlsanity.com/paper/1511.01432/) (2015)
- [Generating Videos with Dynamics-aware Implicit Generative Adversarial Networks](https://mlsanity.com/paper/2202.10571/) (2022)
- [Learning Fast Samplers for Diffusion Models by Differentiating Through Sample Quality](https://mlsanity.com/paper/2202.05830/) (2022)
- [A Good Image Generator Is What You Need for High-Resolution Video Synthesis](https://mlsanity.com/paper/2104.15069/) (2021)
- [Diffusion Models for Video Prediction and Infilling](https://mlsanity.com/paper/2206.07696/) (2022)
- [Train Sparsely, Generate Densely: Memory-efficient Unsupervised Training of High-resolution Temporal GAN](https://mlsanity.com/paper/1811.09245/) (2018)
- [FlowBot3D: Learning 3D Articulation Flow to Manipulate Articulated Objects](https://mlsanity.com/paper/2205.04382/) (2022)
- [TAX-Pose: Task-Specific Cross-Pose Estimation for Robot Manipulation](https://mlsanity.com/paper/2211.09325/) (2022)

## Built on by (most cited first, 9 of 9)

- [CHAMP: Conformalized 3D Human Multi-Hypothesis Pose Estimators](https://mlsanity.com/paper/2407.06141/) (2024)
- [Learning Correspondence for Deformable Objects](https://mlsanity.com/paper/2405.08996/) (2024)
- [Safe Deep Model-Based Reinforcement Learning with Lyapunov Functions](https://mlsanity.com/paper/2405.16184/) (2024)
- [CUPS: Improving Human Pose-Shape Estimators with Conformalized Deep Uncertainty](https://mlsanity.com/paper/2412.10431/) (2024)
- [GeoPE:A Unified Geometric Positional Embedding for Structured Tensors](https://mlsanity.com/paper/2512.04963/) (2025)
- [H2OFlow: Grounding Human-Object Affordances with 3D Generative Models and Dense Diffused Flows](https://mlsanity.com/paper/2510.21769/) (2025)
- [Towards Improving Learning from Demonstration Algorithms via MCMC Methods](https://mlsanity.com/paper/2405.02243/) (2024)
- [Theoretical Analysis for Expectation-Maximization-Based Multi-Model 3D Registration](https://mlsanity.com/paper/2405.08991/) (2024)
- [Rocket Landing Control with Grid Fins and Path-following using MPC](https://mlsanity.com/paper/2405.16191/) (2024)

## Abstract (arXiv)

Diffusion models have exhibited promising progress in video generation. However, they often struggle to retain consistent details within local regions across frames. One underlying cause is that traditional diffusion models approximate Gaussian noise distribution by utilizing predictive noise, without fully accounting for the impact of inherent information within the input itself. Additionally, these models emphasize the distinction between predictions and references, neglecting information intrinsic to the videos. To address this limitation, inspired by the self-attention mechanism, we propose a novel text-to-video (T2V) generation network structure based on diffusion models, dubbed Additional Perturbation for Latent noise with Adversarial training (APLA). Our approach only necessitates a single video as input and builds upon pre-trained stable diffusion networks. Notably, we introduce an additional compact network, known as the Video Generation Transformer (VGT). This auxiliary component is designed to extract perturbations from the inherent information contained within the input, thereby refining inconsistent pixels during temporal predictions. We leverage a hybrid architecture of transformers and convolutions to compensate for temporal intricacies, enhancing consistency between different frames within the video. Experiments demonstrate a noticeable improvement in the consistency of the generated videos both qualitatively and quantitatively.