STRIDE: Training-Free Diversity Guidance via PCA-Directed Feature Perturbation in Single-Step Diffusion Models

• URL: http://arxiv.org/abs/2605.11494v1
• Date: Tue, 12 May 2026 04:10:42 GMT
• Title: STRIDE: Training-Free Diversity Guidance via PCA-Directed Feature Perturbation in Single-Step Diffusion Models
• Abstract summary: We argue that effective diversity injection in few-step models requires perturbations that respect the model's learned feature geometry.<n>We propose STRIDE, a training-free and optimization-free method that operates in a single forward pass.
• Score: 29.580811134498987
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Distilled one-step (T=1) or few-step (T$\leq$4) diffusion models enable real-time image generation but often exhibit reduced sample diversity compared to their multi-step counterparts. In multi-step diffusion, diversity can be introduced through schedules, trajectories, or iterative optimization; however, these mechanisms are unavailable in the few-step or single-step setting, limiting the effectiveness of existing diversity-enhancing methods. A natural alternative is to perturb intermediate features, but naive feature perturbation is often ineffective, either yielding limited diversity gains or degrading generation quality. We argue that effective diversity injection in few-step models requires perturbations that respect the model's learned feature geometry. Based on this insight, we propose STRIDE, a training-free and optimization-free method that operates in a single forward pass. STRIDE injects spatially coherent (pink) noise into intermediate transformer features, projected onto the principal components of the model's own activations, ensuring that perturbations lie on the learned feature manifold. This design enables controlled variation along meaningful directions in the representation space. Extensive experiments on FLUX.1-schnell and SD3.5 Turbo across COCO, DrawBench, PartiPrompts, and GenEval show that STRIDE consistently improves diversity while maintaining strong text alignment. In particular, STRIDE reduces intra-batch similarity with minimal impact on CLIP score, and Pareto-dominates existing training-free baselines on the diversity-fidelity frontier. These results highlight that, in the absence of iterative refinement, improving diversity in few-step and one-step diffusion depends not on increasing perturbation strength, but on aligning perturbations with the model's internal representation structure.

Related papers

• Don't Settle at the Mode! Mitigating Diversity Collapse in Pretrained Flow Models via Feature Self-Guidance [26.478519452337434]
  State-of-the-art flow models generate stunning images from text or image prompts.<n>They suffer from diversity collapse when generating multiple samples under the same conditioning.<n>We introduce an efficient, training-free self-guidance mechanism to mitigate diversity collapse without requiring additional reward models.
  arXiv  Detail & Related papers  (2026-06-25T17:59:47Z)
• Initialization is Half the Battle: Generating Diverse Images from a Guidance Potential Posterior [51.62608628984139]
  We formulate selecting the initial noise from a guidance potential posterior, which effectively re-weights the prior towards diversity-rich regions.<n>To sample from this distribution efficiently, we introduce Diversity-inducing Initialization (DivIn)<n>Our method serves as an inference-time diversity enhancement compatible with both diffusion and flow matching models.
  arXiv  Detail & Related papers  (2026-06-01T16:25:10Z)
• Training-Free Generation of Diverse and High-Fidelity Images via Prompt Semantic Space Optimization [50.5332987313297]
  We propose Token-Prompt embedding Space Optimization (TPSO), a training-free and model-agnostic module.<n>TPSO introduces learnable parameters to explore underrepresented regions of the token embedding space, reducing the tendency of the model to repeatedly generate samples from strong modes of the learned distribution.<n>In experiments on MS-COCO and three diffusion backbones, TPSO significantly enhances generative diversity, improving baseline performance from 1.10 to 4.18 points, without sacrificing image quality.
  arXiv  Detail & Related papers  (2025-11-25T00:42:09Z)
• Diversity Has Always Been There in Your Visual Autoregressive Models [78.27363151940996]
  Visual Autoregressive ( VAR) models have recently garnered significant attention for their innovative next-scale prediction paradigm.<n>Despite their efficiency, VAR models often suffer from the diversity collapse, analogous to that observed in few-step distilled diffusion models.<n>We introduce Diverse VAR, a simple yet effective approach that restores the generative diversity of VAR models without requiring any additional training.
  arXiv  Detail & Related papers  (2025-11-21T09:24:09Z)
• Hybrid Autoregressive-Diffusion Model for Real-Time Sign Language Production [0.0]
  We develop a hybrid approach that combines autoregressive and diffusion models for Sign Language Production (SLP)<n>To capture fine-grained body movements, we design a Multi-Scale Pose Representation module that separately extracts detailed features from distinct articulators.<n>We introduce a Confidence-Aware Causal Attention mechanism that utilizes joint-level confidence scores to dynamically guide the pose generation process.
  arXiv  Detail & Related papers  (2025-07-12T01:34:50Z)
• Constrained Discrete Diffusion [61.81569616239755]
  This paper introduces Constrained Discrete Diffusion (CDD), a novel integration of differentiable constraint optimization within the diffusion process.<n>CDD directly imposes constraints into the discrete diffusion sampling process, resulting in a training-free and effective approach.
  arXiv  Detail & Related papers  (2025-03-12T19:48:12Z)
• One-Step Diffusion Model for Image Motion-Deblurring [85.76149042561507]
  We propose a one-step diffusion model for deblurring (OSDD), a novel framework that reduces the denoising process to a single step.<n>To tackle fidelity loss in diffusion models, we introduce an enhanced variational autoencoder (eVAE), which improves structural restoration.<n>Our method achieves strong performance on both full and no-reference metrics.
  arXiv  Detail & Related papers  (2025-03-09T09:39:57Z)
• Supercharged One-step Text-to-Image Diffusion Models with Negative Prompts [19.609393551644562]
  We introduce textbfNegative-textbfAway textbfSteer textbfAttention (NASA), an efficient method that integrates negative prompts into one-step diffusion models.<n>NASA operates within the intermediate representation space by leveraging cross-attention mechanisms to suppress undesired visual attributes.
  arXiv  Detail & Related papers  (2024-12-03T18:56:32Z)
• Few-Shot Image Generation by Conditional Relaxing Diffusion Inversion [37.18537753482751]
  Conditional Diffusion Relaxing Inversion (CRDI) is designed to enhance distribution diversity in synthetic image generation. CRDI does not rely on fine-tuning based on only a few samples. It focuses on reconstructing each target image instance and expanding diversity through few-shot learning.
  arXiv  Detail & Related papers  (2024-07-09T21:58:26Z)
• Ensembling Diffusion Models via Adaptive Feature Aggregation [14.663257610094625]
  Leveraging multiple high-quality models to produce stronger generation ability is valuable, but has not been extensively studied.<n>Existing methods primarily adopt parameter merging strategies to produce a new static model.<n>We propose Adaptive Feature Aggregation (AFA), which dynamically adjusts the contributions of multiple models at the feature level according to various states.
  arXiv  Detail & Related papers  (2024-05-27T11:55:35Z)
• Mitigating Shortcut Learning with Diffusion Counterfactuals and Diverse Ensembles [104.60508550106618]
  We propose DiffDiv, an ensemble diversification framework exploiting Diffusion Probabilistic Models (DPMs)<n>We show that DPMs can generate images with novel feature combinations, even when trained on samples displaying correlated input features.<n>We show that DPM-guided diversification is sufficient to remove dependence on shortcut cues, without a need for additional supervised signals.
  arXiv  Detail & Related papers  (2023-11-23T15:47:33Z)
• Auto-regressive Image Synthesis with Integrated Quantization [55.51231796778219]
  This paper presents a versatile framework for conditional image generation. It incorporates the inductive bias of CNNs and powerful sequence modeling of auto-regression. Our method achieves superior diverse image generation performance as compared with the state-of-the-art.
  arXiv  Detail & Related papers  (2022-07-21T22:19:17Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.