# Why Are Conditional Generative Models Better Than Unconditional Ones?

arXiv: https://arxiv.org/abs/2212.00362 · HTML: https://mlsanity.com/paper/2212.00362/

## Key facts

- Cited by 28 papers in the map (data through 14 Oct 2026).
- Placed in Diffusion foundations and sampling, one of 14 directions in Diffusion models.
- Builds on 16 of its 22 references that are in the map.
- First posted on arXiv 1 Dec 2022.

## Placed in

- Diffusion models / [Diffusion foundations and sampling](https://mlsanity.com/diffusion/diffusion-foundations-and-sampling/)

## Builds on (most cited first, 16 of 16)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [A Simple Framework for Contrastive Learning of Visual Representations](https://mlsanity.com/paper/2002.05709/) (2020)
- [Score-Based Generative Modeling through Stochastic Differential Equations](https://mlsanity.com/paper/2011.13456/) (2020)
- [Diffusion Models Beat GANs on Image Synthesis](https://mlsanity.com/paper/2105.05233/) (2021)
- [Momentum Contrast for Unsupervised Visual Representation Learning](https://mlsanity.com/paper/1911.05722/) (2019)
- [Improved Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2102.09672/) (2021)
- [Deep Learning Face Attributes in the Wild](https://mlsanity.com/paper/1411.7766/) (2014)
- [Large Scale GAN Training for High Fidelity Natural Image Synthesis](https://mlsanity.com/paper/1809.11096/) (2018)
- [Improved Baselines with Momentum Contrastive Learning](https://mlsanity.com/paper/2003.04297/) (2020)
- [DPM-Solver: A Fast ODE Solver for Diffusion Probabilistic Model Sampling in Around 10 Steps](https://mlsanity.com/paper/2206.00927/) (2022)
- [An Empirical Study of Training Self-Supervised Vision Transformers](https://mlsanity.com/paper/2104.02057/) (2021)
- [Maximum Likelihood Training of Score-Based Diffusion Models](https://mlsanity.com/paper/2101.09258/) (2021)
- [Analytic-DPM: an Analytic Estimate of the Optimal Reverse Variance in Diffusion Probabilistic Models](https://mlsanity.com/paper/2201.06503/) (2022)
- [Instance-Conditioned GAN](https://mlsanity.com/paper/2109.05070/) (2021)
- [Triple Generative Adversarial Nets](https://mlsanity.com/paper/1703.02291/) (2017)
- [Self-labeled Conditional GANs](https://mlsanity.com/paper/2012.02162/) (2020)

## Built on by (most cited first, 28 of 28)

- [GenAD: Generalized Predictive Model for Autonomous Driving](https://mlsanity.com/paper/2403.09630/) (2024)
- [Return of Unconditional Generation: A Self-supervised Representation Generation Method](https://mlsanity.com/paper/2312.03701/) (2023)
- [SlotDiffusion: Object-Centric Generative Modeling with Diffusion Models](https://mlsanity.com/paper/2305.11281/) (2023)
- [A Survey on Diffusion Models for Time Series and Spatio-Temporal Data](https://mlsanity.com/paper/2404.18886/) (2024)
- [Proteina: Scaling Flow-based Protein Structure Generative Models](https://mlsanity.com/paper/2503.00710/) (2025)
- [Trajectory Consistency Distillation: Improved Latent Consistency Distillation by Semi-Linear Consistency Function with Trajectory Mapping](https://mlsanity.com/paper/2402.19159/) (2024)
- [Diffusion Models and Semi-Supervised Learners Benefit Mutually with Few Labels](https://mlsanity.com/paper/2302.10586/) (2023)
- [Toward Understanding Generative Data Augmentation](https://mlsanity.com/paper/2305.17476/) (2023)
- [DisCo-Diff: Enhancing Continuous Diffusion Models with Discrete Latents](https://mlsanity.com/paper/2407.03300/) (2024)
- [Highly Compressed Tokenizer Can Generate Without Training](https://mlsanity.com/paper/2506.08257/) (2025)
- [Visual Chain-of-Thought Diffusion Models](https://mlsanity.com/paper/2303.16187/) (2023)
- [MotionMix: Weakly-Supervised Diffusion for Controllable Motion Generation](https://mlsanity.com/paper/2401.11115/) (2024)
- [What Happens Next? Anticipating Future Motion by Generating Point Trajectories](https://mlsanity.com/paper/2509.21592/) (2025)
- [Guided Diffusion from Self-Supervised Diffusion Features](https://mlsanity.com/paper/2312.08825/) (2023)
- [Cross-Domain Graph Data Scaling: A Showcase with Diffusion Models](https://mlsanity.com/paper/2406.01899/) (2024)
- [Versatile Cardiovascular Signal Generation with a Unified Diffusion Transformer](https://mlsanity.com/paper/2505.22306/) (2025)
- [Rethinking cluster-conditioned diffusion models for label-free image synthesis](https://mlsanity.com/paper/2403.00570/) (2024)
- [Masked Generative Modeling with Enhanced Sampling Scheme](https://mlsanity.com/paper/2309.07945/) (2023)
- [Conditioning Residuals for Diffusion Models via Representation Feedback](https://mlsanity.com/paper/2505.10999/) (2025)
- [Compositional Discrete Latent Code for High Fidelity, Productive Diffusion Models](https://mlsanity.com/paper/2507.12318/) (2025)
- [Rethinking Diffusion Model in High Dimension](https://mlsanity.com/paper/2503.08643/) (2025)
- [Scaling Up Deep Clustering Methods Beyond ImageNet-1K](https://mlsanity.com/paper/2406.01203/) (2024)
- [Leveraging Discriminative Latent Representations for Conditioning GAN-Based Speech Enhancement](https://mlsanity.com/paper/2508.20859/) (2025)
- [A Hybrid Conditional Diffusion-DeepONet Framework for High-Fidelity Stress Prediction in Hyperelastic Materials](https://mlsanity.com/paper/2603.18225/) (2026)
- [CAFusion: Controllable Anatomical Synthesis of Perirectal Lymph Nodes via SDF-guided Diffusion](https://mlsanity.com/paper/2503.06919/) (2025)
- [Challenges and Limitations of Generative AI in Synthesizing Wearable Sensor Data](https://mlsanity.com/paper/2505.14206/) (2025)
- [Flow Matching with Optimized Subclass Priors for Medical Image Augmentation](https://mlsanity.com/paper/2605.16469/) (2026)
- [From Pixels, Without Pre-training: Joint Generative and Self-Supervised Representation Learning in One Model](https://mlsanity.com/paper/2610.05711/) (2026)

## Abstract (arXiv)

Extensive empirical evidence demonstrates that conditional generative models are easier to train and perform better than unconditional ones by exploiting the labels of data. So do score-based diffusion models. In this paper, we analyze the phenomenon formally and identify that the key of conditional learning is to partition the data properly. Inspired by the analyses, we propose self-conditioned diffusion models (SCDM), which is trained conditioned on indices clustered by the k-means algorithm on the features extracted by a model pre-trained in a self-supervised manner. SCDM significantly improves the unconditional model across various datasets and achieves a record-breaking FID of 3.94 on ImageNet 64x64 without labels. Besides, SCDM achieves a slightly better FID than the corresponding conditional model on CIFAR10.