mlsanityData to 14 Oct 2026

Training-Free Guidance Beyond Differentiability: Scalable Path Steering with Tree Search in Diffusion and Flow Models

arXiv 2502.11420 ↗First posted 17 Feb 2025cs.LGcited by 24

Not on the map. Why?

Abstract (arXiv)

Training-free guidance enables controlled generation in diffusion and flow models, but most methods rely on gradients and assume differentiable objectives. This work focuses on training-free guidance addressing challenges from non-differentiable objectives and discrete data distributions. We propose TreeG: Tree Search-Based Path Steering Guidance, applicable to both continuous and discrete settings in diffusion and flow models. TreeG offers a unified framework for training-free guidance by proposing, evaluating, and selecting candidates at each step, enhanced with tree search over active paths and parallel exploration. We comprehensively investigate the design space of TreeG over the candidate proposal module and the evaluation function, instantiating TreeG into three novel algorithms. Our experiments show that TreeG consistently outperforms top guidance baselines in symbolic music generation, small molecule design, and enhancer DNA design with improvements of 29.01%, 16.6%, and 18.43%. Additionally, we identify an inference-time scaling law showing TreeG's scalability in inference-time computation.

Key facts. Cited by 24 papers in the map (data through 14 Oct 2026). Builds on 65 of its 87 references that are in the map. First posted on arXiv 17 Feb 2025.

Builds on

65 of 87 references.

References · 65

  1. 2020Denoising Diffusion Probabilistic Modelscited by 21,076
  2. 2020Score-Based Generative Modeling through Stochastic Differential Equationscited by 9,122
  3. 2021Diffusion Models Beat GANs on Image Synthesiscited by 8,255
  4. 2022Classifier-Free Diffusion Guidancecited by 5,847
  5. 2019Generative Modeling by Estimating Gradients of the Data Distributioncited by 4,101
  6. 2023Diffusion Policy: Visuomotor Policy Learning via Action Diffusioncited by 2,838
  7. 2016Categorical Reparameterization with Gumbel-Softmaxcited by 2,739
  8. 2020Learning to summarize from human feedbackcited by 2,588
  9. 2021WebGPT: Browser-assisted question-answering with human feedbackcited by 1,745
  10. 2021Structured Denoising Diffusion Models in Discrete State-Spacescited by 1,719
  11. 2022Diffusion Posterior Sampling for General Noisy Inverse Problemscited by 1,416
  12. 2023Training Diffusion Models with Reinforcement Learningcited by 982
38 more
  1. 2023Diffusion Model Alignment Using Direct Preference Optimizationcited by 790
  2. 2023Discrete Diffusion Modeling by Estimating the Ratios of the Data Distributioncited by 745
  3. 2021Argmax Flows and Multinomial Diffusion: Learning Categorical Distributionscited by 574
  4. 2022DiGress: Discrete Denoising diffusion for graph generationcited by 520
  5. 2022Is Conditional Generative Modeling all you need for Decision-Making?cited by 516
  6. 2022A Continuous Time Framework for Discrete Denoising Modelscited by 451
  7. 2023Universal Guidance for Diffusion Modelscited by 437
  8. 2024Discrete Flow Matchingcited by 317
  9. 2024Generative Flows on Discrete State-Spaces: Enabling Multimodal Flows with Applications to Protein Co-Designcited by 305
  10. 2024Flow Matching Guide and Codecited by 301
  11. 2023FreeDoM: Training-Free Energy-Guided Conditional Diffusion Modelcited by 294
  12. 2023Aligning Text-to-Image Diffusion Models with Reward Backpropagationcited by 288
  13. 2021Therapeutics Data Commons: Machine Learning Datasets and Tasks for Drug Discovery and Developmentcited by 264
  14. 2025Inference-Time Scaling for Diffusion Models beyond Scaling Denoising Stepscited by 234
  15. 2024Adjoint Matching: Fine-tuning Flow and Diffusion Generative Models with Memoryless Stochastic Optimal Controlcited by 233
  16. 2023Practical and Asymptotically Exact Conditional Sampling in Diffusion Modelscited by 232
  17. 2025A General Framework for Inference-time Scaling and Steering of Diffusion Modelscited by 224
  18. 2018Enabling Factorized Piano Music Modeling and Generation with the MAESTRO Datasetcited by 217
  19. 2022Score-based Continuous-time Discrete Diffusion Modelscited by 169
  20. 2023Manifold Preserving Guided Diffusioncited by 157
  21. 2022Sample Efficiency Matters: A Benchmark for Practical Molecular Optimizationcited by 155
  22. 2024Fine-Tuning of Continuous-Time Diffusion Models as Entropy-Regularized Controlcited by 143
  23. 2024Derivative-Free Guidance in Continuous and Discrete Diffusion Models with Soft Value-Based Decodingcited by 136
  24. 2024Unlocking Guidance for Discrete State-Space Diffusion and Flow Modelscited by 129
  25. 2024Theoretical guarantees on the best-of-n alignment policycited by 124
  26. 2024D-Flow: Differentiating through Flows for Controlled Generationcited by 122
  27. 2024Dirichlet Flow Matching with Applications to DNA Sequence Designcited by 115
  28. 2021Compound Word Transformer: Learning to Compose Full-Song Music over Dynamic Directed Hypergraphscited by 110
  29. 2023End-to-End Diffusion Latent Optimization Improves Classifier Guidancecited by 108
  30. 2024TFG: Unified Training-Free Guidance for Diffusion Modelscited by 107
  31. 2023Optimizing Diffusion Noise Can Serve As Universal Motion Priorscited by 98
  32. 2024Understanding Reinforcement Learning-Based Fine-Tuning of Diffusion Models: A Tutorial and Reviewcited by 98
  33. 2025Inference-Time Alignment in Diffusion Models with Reward-Guided Generation: Tutorial and Reviewcited by 88
  34. 2021Guided-TTS: A Diffusion Model for Text-to-Speech via Classifier Guidancecited by 84
  35. 2020POP909: A Pop-song Dataset for Music Arrangement Generationcited by 83
  36. 2024Fine-Tuning Discrete Diffusion Models via Reward Optimization with Applications to DNA and Protein Designcited by 79
  37. 2023Dirichlet Diffusion Score Model for Biological Sequence Generationcited by 75
  38. 2024Amortizing intractable inference in diffusion models for vision, language, and controlcited by 72

50 most cited of 65 All 65 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 10; 2026: 14.

24 papers.

Sort:

By area:

  1. 2025Inference-time Scaling of Diffusion Models through Classical Searchcited by 49
  2. 2025Dynamic Search for Inference-Time Alignment in Diffusion Modelscited by 37
  3. 2025Feedback Guidance of Diffusion Modelscited by 16
  4. 2025HardFlow: Hard-Constrained Sampling for Flow-Matching Models via Trajectory Optimizationcited by 11
  5. 2026Training-Free Adaptation of Diffusion Models via Doob's h-Transformcited by 11
  6. 2025Inference-Time Scaling of Diffusion Language Models via Trajectory Refinementcited by 7
  7. 2025AutoRefiner: Improving Autoregressive Video Diffusion Models via Reflective Refinement Over the Stochastic Sampling Pathcited by 4
  8. 2026Entropy Aware Reward Guidance for Diffusion Language Model Alignmentcited by 3
  9. 2025Tree Reward-Aligned Search for TReASURe in Masked Diffusion Language Modelscited by 2
  10. 2025Local MAP Sampling for Diffusion Modelscited by 2
  11. 2026Inference-Time Scaling in Diffusion Models through Iterative Partial Refinementcited by 2
  12. 2026Search-Augmented Masked Diffusion Models for Constrained Generationcited by 1
12 more
  1. 2026Slowly Annealed Langevin Dynamics: Theory and Applications to Training-Free Guided Generationcited by 1
  2. 2026BFMT: Enhancing Search Capabilities of Tree Sampler via Bootstrap Flow-Map Treecited by 1
  3. 2026VGAS: Variance-Reduced Guidance and Adaptive Selection for Training-Free Reward Alignment in Discrete Diffusioncited by 1
  4. 2025Guiding Diffusion with Deep Geometric Moments: Balancing Fidelity and Variationcited by 0
  5. 2025Iterative Inference-time Scaling with Adaptive Frequency Steering for Image Super-Resolutioncited by 0
  6. 2026Flow-Direct: Feedback-Efficient and Reusable Guidance for Flow Models via Non-Parametric Guidance Fieldcited by 0
  7. 2026Histogram-constrained Image Generationcited by 0
  8. 2026Feature-Space Guided Diffusion for Realistic Ultrasound Image Synthesiscited by 0
  9. 2026LC-GRPO: Bridging Train-Inference Gap for Flow-Based GRPO with Langevin Correctioncited by 0
  10. 2026Language-Informed Flow Matching for Trend-Guided Structure-Based 3D Molecular Generationcited by 0
  11. 2026LATS: Levy Adaptive Tree Sampling for Feedback-Driven Diverse Target Discoverycited by 0
  12. 2026RIDE: Reference-Anchored Inference-Time Diffusion Editing for Scaffold Hoppingcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026Plug-and-Play Guidance for Discrete Diffusion Models via Gradient-Informed Logit Correction
  2. 2025TFG-Flow: Training-free Guidance in Multimodal Generative Flow
  3. 2024Training-Free Guidance for Discrete Diffusion Models for Molecular Generation
  4. 2025Tortoise and Hare Guidance: Accelerating Diffusion Model Inference with Multirate Integration
  5. 2026Opening the Black Box of Classifier-Free Guidance via Information Bottleneck
  6. 2025Token Perturbation Guidance for Diffusion Models
  7. 2026Analytic Distribution of Classifier-Free Guidance for Schedule Design
  8. 2024Simple Guidance Mechanisms for Discrete Diffusion Models