mlsanityData to 14 Oct 2026

Efficient Controllable Diffusion via Optimal Classifier Guidance

arXiv 2505.21666 ↗First posted 27 May 2025cs.LG cs.AIcited by 2

On the map: Diffusion models → Diffusion foundations and sampling

Abstract (arXiv)

The controllable generation of diffusion models aims to steer the model to generate samples that optimize some given objective functions. It is desirable for a variety of applications including image generation, molecule generation, and DNA/sequence generation. Reinforcement Learning (RL) based fine-tuning of the base model is a popular approach but it can overfit the reward function while requiring significant resources. We frame controllable generation as a problem of finding a distribution that optimizes a KL-regularized objective function. We present SLCD -- Supervised Learning based Controllable Diffusion, which iteratively generates online data and trains a small classifier to guide the generation of the diffusion model. Similar to the standard classifier-guided diffusion, SLCD's key computation primitive is classification and does not involve any complex concepts from RL or control. Via a reduction to no-regret online learning analysis, we show that under KL divergence, the output from SLCD provably converges to the optimal solution of the KL-regularized objective. Further, we empirically demonstrate that SLCD can generate high quality samples with nearly the same inference time as the base model in both image generation with continuous diffusion and biological sequence generation with discrete diffusion. Our code is available at https://github.com/Owen-Oertell/slcd

Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Placed in Diffusion foundations and sampling, one of 14 directions in Diffusion models. Builds on 34 of its 47 references that are in the map. First posted on arXiv 27 May 2025.

Builds on

34 of 47 references.

Closest ancestors · 10

Same direction, newest first

  1. 2025Convergence Of Consistency Model With Multistep Sampling Under General Data Assumptions6 May
  2. 2024Derivative-Free Guidance in Continuous and Discrete Diffusion Models with Soft Value-Based Decoding15 Aug
  3. 2022Improved Analysis of Score-based Generative Modeling: User-Friendly Bounds under Minimal Smoothness Assumptions3 Nov
  4. 2022Sampling is as easy as learning the score: theory for diffusion models with minimal data assumptions22 Sep
  5. 2022Classifier-Free Diffusion Guidance26 Jul
  6. 2021High-Resolution Image Synthesis with Latent Diffusion Models20 Dec
  7. 2021Diffusion Models Beat GANs on Image Synthesis11 May
  8. 2020Score-Based Generative Modeling through Stochastic Differential Equations26 Nov
  9. 2020Denoising Diffusion Implicit Models6 Oct
  10. 2020Denoising Diffusion Probabilistic Models19 Jun

Other references · 24

  1. 2017Proximal Policy Optimization Algorithmscited by 15,215
  2. 2022Photorealistic Text-to-Image Diffusion Models with Deep Language Understandingcited by 6,355
  3. 2022Imagen Video: High Definition Video Generation with Diffusion Modelscited by 1,778
  4. 2022Diffusion Posterior Sampling for General Noisy Inverse Problemscited by 1,416
  5. 2024Simple and Effective Masked Diffusion Language Modelscited by 848
  6. 2022Equivariant Diffusion for Molecule Generation in 3Dcited by 677
  7. 2023DPOK: Reinforcement Learning for Fine-tuning Text-to-Image Diffusion Modelscited by 450
  8. 2023Directly Fine-Tuning Diffusion Models on Differentiable Rewardscited by 448
  9. 2023Aligning Text-to-Image Diffusion Models with Reward Backpropagationcited by 288
  10. 2017Gate-Variants of Gated Recurrent Unit (GRU) Neural Networkscited by 242
  11. 2022Diffusion probabilistic modeling of protein backbones in 3D for the motif-scaffolding problemcited by 236
  12. 2024Adjoint Matching: Fine-tuning Flow and Diffusion Generative Models with Memoryless Stochastic Optimal Controlcited by 233
12 more
  1. 2023Practical and Asymptotically Exact Conditional Sampling in Diffusion Modelscited by 232
  2. 2021The Statistical Complexity of Interactive Decision Makingcited by 218
  3. 2024Fine-Tuning of Continuous-Time Diffusion Models as Entropy-Regularized Controlcited by 143
  4. 2023Controlled Decoding from Language Modelscited by 135
  5. 2024Unlocking Guidance for Discrete State-Space Diffusion and Flow Modelscited by 129
  6. 2014Reinforcement and Imitation Learning via Interactive No-Regret Learningcited by 118
  7. 2017Deeply AggreVaTeD: Differentiable Imitation Learning for Sequential Predictioncited by 93
  8. 2024Feedback Efficient Online Fine-Tuning of Diffusion Modelscited by 53
  9. 2019Online Non-Convex Learning: Following the Perturbed Leader is Optimalcited by 32
  10. 2025Qsharp: Provably Optimal Distributional RL for LLM Post-Trainingcited by 15
  11. 2024RL for Consistency Models: Faster Reward Guided Text-to-Image Generationcited by 12
  12. 2024The Central Role of the Loss Function in Reinforcement Learningcited by 8

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 2.

2 papers.

  1. 2026Do Less, Achieve More: Do We Need Every-Step Optimization for RL Fine-tuning of Diffusion Models?cited by 6
  2. 2026Provably Reliable Classifier Guidance via Cross-Entropy Controlcited by 1

Similar, not cited

Close in wording, no citation either way

  1. 2024Simple Guidance Mechanisms for Discrete Diffusion Models
  2. 2023Towards Controllable Diffusion Models via Reward-Guided Exploration
  3. 2023Elucidating The Design Space of Classifier-Guided Diffusion Generation
  4. 2026Spectral Guidance for Flexible and Efficient Control of Diffusion Models
  5. 2025Towards a unified framework for guided diffusion models
  6. 2026Diffusion Controller: Framework, Algorithms and Parameterization
  7. 2024Understanding Reinforcement Learning-Based Fine-Tuning of Diffusion Models: A Tutorial and Review
  8. 2025CCS: Controllable and Constrained Sampling with Diffusion Models via Initial Noise Perturbation