Diffusion Alignment

Follow this topic Ask about this topic

Momentum

4 papers in the last four weeks, against 1 the four weeks before. 0.0% of all new papers.

Jul 6Week of Sep 21

Latest papers 65

Oct 1, 2026cs.LG

iADD: Improving Alignment and Diversity in Diffusion Policy Optimization

Reinforcement learning based post training of diffusion models, such as Denoising Diffusion Policy Optimization (DDPO), optimizes a reverse diffusion process under a reward function. However, current approaches to reward optimizations do so at the cost of diversity and quality. In this paper, we provide better tradeoffs through careful theoretical considerations and method design. We analyze the theoretical framework and mathematically demonstrate that \emph{only-latter timestep} updates of diffusion model may be harmful for diversity contrary to the conclusions presented in a previous work. Additionally, we propose an incremental Feynman-Kac training based on strong theoretical foundations in order to achieve the best-yet alignment-diversity tradeoffs. We perform extensive experiments and compare our method against related diffusion policy optimization approaches in three different tasks and also provide strong ablations for each component, thus validating strong performance gains in both alignment and diversity.

Ashok Prasad Neupane, Saugat Adhikari, Pramish Paudel +2

Pulchowk Campus, IOE, Tribhuvan University, Lalitpur, Nepal · INSAIT, Sofia University “St. Kliment Ohridski”, Bulgaria · NAAMII, Kathmandu, Nepal

Diffusion Alignment Diversity Reward Functions

Sep 30, 2026cs.CV

Just Align $\bm{x}$ : Aligning Predictions, Not Representations

Representation alignment has become an effective way to accelerate diffusion training, but its benefits do not transfer reliably to pixel-space clean-image prediction. In JiT, we find that auxiliary feature alignment can improve access to semantic features while reducing access to image variation needed for clean-image prediction, creating a mismatch between the auxiliary objective and the denoising task. This suggests a different principle: auxiliary supervision should improve the prediction target itself rather than impose a separate representation target. We introduce JAx (Just Align x), a prediction-supervision method that aligns clean-image predictions across noise levels. JAx couples a noisier student observation with a cleaner observation through a Markov degradation that preserves the original JiT input distribution. Under this coupling, the oracle prediction from the cleaner state has the same conditional mean as the optimal JiT target, while its conditional target covariance is no greater. Thus, oracle prediction alignment preserves the population JiT objective up to a constant while providing a lower-variance training target. To make this construction practical with an imperfect EMA teacher, JAx combines ground-truth supervision with a reliability-gated coupling band that selects nearby teacher states based on prediction risk. On ImageNet 256x256, JAx consistently improves FID and accelerates convergence across JiT-B/16, L/16, and H/16, without an external encoder or changes to the architecture or sampling procedure. Gradient diagnostics further show reduced minibatch gradient variance, while ablations demonstrate that the gains cannot be explained by time reweighting alone. These results show that prediction-space supervision provides a simple and principled alternative to representation alignment for pixel-space generative models.

Yuyao Zhang, Yuwei Hu, Ziyang Mai +1

Dartmouth College

Diffusion Alignment Diffusion Models Encoders

Sep 30, 2026math.ST

WEIRDO: WEak resIdual Regularized DOob's h-transform diffusion alignment

We study the problem of estimating the guidance that steers the distribution learned by a diffusion generative model toward a tilted target

q_0 \propto w\,p_0

at inference time. Relying on the stochastic optimal control approach, we observe that the exact drift correction is the gradient of the logarithm of Doob's

h

-function, and we study the problem of estimating it from a sample. In the present paper, we assume that the score of the pretrained model is available, that the tilting weight is bounded and positive, and that the reference distribution has a bounded support, no smoothness of the weight is required. Introducing a penalized least-squares risk in which the penalty is the residual of the space-time harmonicity equation satisfied by the

h

-function, measured in a dual Sobolev norm, we derive high-probability bounds on the squared error of the resulting guidance estimate. Since the penalty vanishes at the target, the estimator is free of regularization bias, and in favourable scenarios its rate of convergence is faster than the minimax rate of estimating first-order derivatives of a smooth regression function. Assuming that

w

is bounded and positive with

\mathbb{E}_{p_0}[w^{-\mathrm{s}}] < \infty

for some

\mathrm{s} \in (0,\infty]

, and that the reference data are compactly supported, we prove that the guidance is estimable in squared

L^2

at rate

\varepsilon_n^{\mathrm{s}/(\mathrm{s}+4)}

, where

\varepsilon_n = n^{-2(β-1)/(2(β-1)+d)}.

We also transfer the obtained bounds to the total variation distance between the marginals of the estimated and the exactly guided samplers, and illustrate the performance of the suggested approach with numerical experiments.

Denis Suchkov

HSE University, Russian Federation

Diffusion Alignment Wasserstein Gradient Flows Regularization Guidance Stochastic Optimal Control Generative Models

Sep 30, 2026cs.LG

Manifold-Constrained Initial Noise Optimization for Efficient Generative Model Alignment

Recent advances in distillation and flow-map models have enabled deterministic one- or few-step generation for high-quality data, facilitating a new branch of reward alignment approaches that directly optimize the initial noise from a Gaussian distribution. However, most existing initial-noise optimization methods rely on first-order gradient information, which is either inapplicable or suffers from instability and inefficiency in black-box reward scenarios. Here, we introduce ZeNOVA, a stable and efficient initial noise alignment method in a gradient-free manner. Specifically, we address existing algorithms' major challenge in black-box scenarios through annealed soft-value guidance, manifold-constrained hyperspherical Langevin dynamics, and Metropolis-Hastings jumping. Extensive experiments on image and video generative models show that ZeNOVA outperforms all evaluated zeroth-order baselines by optimizing the initial noise toward higher rewards substantially more stably while exploiting the geometry of the Gaussian prior, demonstrating its practical applicability to various black-box reward alignment.

Jinho Chang, Jong Chul Ye

Kim Jaechul Graduate School of AI KAIST Seoul, South Korea

Diffusion Alignment Generative Models Langevin Dynamics

Sep 30, 2026stat.ML

Steepest Guidance: A Practical and Principled Approach to Inference-Time Alignment of Flow and Diffusion-based Models

Inference-time alignment of flow and diffusion-based models is critical for achieving flexible generative modeling. Theoretically, Doob's

h

-transform provides an elegant solution to this problem, and most existing methods are based on this principle. However, in practice, estimating the optimal guidance derived from Doob's

h

-transform at inference time is challenging. To deal with this issue, we regard inference-time alignment as a sequential optimization problem in the space of probability measures and propose a novel framework called Steepest Guidance, based on the principle of maximizing local improvement in the objective. We provide a theoretical analysis of the proposed method and demonstrate its effectiveness through extensive experiments.

Shokichi Takakura, Akifumi Wachi, Rei Higuchi +2

LY Corporation · The University of Tokyo · RIKEN AIP

Diffusion Alignment Diffusion Models Generative Models Guidance Robust Optimization Principles

Sep 29, 2026cs.CV

Adaptive Reward Routing: Dynamic Multi-Reward Optimization for Joint Audio-Video Diffusion via Forward-Process RL

Multi-reward guided reinforcement learning (i.e., RL) offers a promising way to improve joint audio-video diffusion models along several complementary objectives, including modality-specific quality, cross-modal semantic alignment, and temporal synchronization. Its effectiveness, however, depends on two quantities that change during training: where reward-driven updates should act, and how competing rewards should be combined. Existing methods tend to rely on fixed routing and reward weights, failing to track evolving model functions. To address these limitations, we propose Adaptive Reward Routing to jointly adapt update locations and reward coordination during forward-process RL (i.e., DiffusionNFT) of joint audio-video diffusion models. Our method consists of two components. (i) Cross-Modal Influence-Guided Routing (Localizing Updates): We use bidirectional cross-attention responses as an efficient proxy for evolving cross-modal influence, dynamically reweighting token-aware losses and scaling gradients across cross-modal layers without additional model interventions. (ii) Preference-Preserving Modality-Aware Reweighting (Coordinating Rewards): We preserve predefined weights as preference priors and use branch-specific reward-gradient interactions as residual corrections after warm-up. This resolves evolving conflicts without letting dominant rewards suppress weak but essential objectives. Extensive experiments demonstrate consistent improvements in modality quality, semantic consistency, and audio-video synchronization over strong RL baselines. Ablations and mechanism analyses further validate the complementary benefits of adaptive update routing and reward coordination.

Songlin Yang, Xiaotong Zhao, Jiacheng Zhang +5

Tencent Video · The University of Hong Kong

Diffusion-Based Reinforcement Learning Methods Diffusion Alignment Reward Gradients Cross-Modal Group-Based Reinforcement Learning Offline Reinforcement Learning

Sep 28, 2026cs.CV

What Visual Generators Need from Teachers: Rethinking Representation Alignment

Representation alignment speeds up diffusion transformer training by pulling an intermediate block of the model (student) toward features of a frozen pretrained encoder (teacher). Which teacher layer to align, and for how long, is still set by convention, and each alternative costs a training run. We find that alignment helps where the student cannot linearly recover the teacher's features, not where it already resembles them. Since a deep teacher layer is largely predictable from the one below, we isolate what each layer adds, its increment, and measure how much of it an unaligned student recovers. The student fills the teacher's hierarchy from the bottom up and stalls near the top, which we call hierarchy filling: even after 400K steps it recovers almost none of the deepest. The recoverability gap is the unrecovered share of an increment, read from one unaligned checkpoint. In short runs that each align one teacher layer at one block, the gap nearly reproduces their ranking by FID improvement, and CKA, a measure of feature similarity, largely reverses it. Representation Alignment and Recoverability Estimation (RARE) picks the teacher layer with the largest gap before training. During training, it tracks each token's remaining distance to that layer, the online counterpart of the gap, weights tokens by it, and phases out the loss once the average distance stops falling. With SiT-B/2 on ImageNet

256\times256

, RARE reaches an FID of 18.02 without guidance and 4.46 with it, ahead of seven alignment baselines including REPA, iREPA and HASTE. It also trains in 14% fewer GPU-hours than iREPA. Its FID stays below iREPA's across model scales, teachers, datasets and backbones.

Yongcong Wang, Hingchin Chen, Mingyu Fan +6

Central South University · The Hong Kong University of Science and Technology · Tsinghua University +7

Diffusion Alignment Diffusion Transformers Teacher Encoders Visual Generation

Sep 28, 2026cs.LG

FestDPO: Few-step Generator Alignment with Direct Preference Optimization

Few-step generative models can generate high-fidelity samples within a few function evaluations. Despite this efficiency, generated samples may not exhibit desirable properties. When these properties are difficult to encode as an explicit reward function, direct preference optimization (DPO) can align generative models using pairwise preference feedback without training a separate reward model. However, extending DPO to few-step generative models is challenging because few-step generative models are generally implicit, making the likelihood evaluation required by DPO intractable. To address this challenge, we introduce Few-step DPO (FestDPO), an extension of DPO for few-step generative models that leverages nonparametric likelihood estimation from empirical samples. By exploiting the fast sampling capabilities of few-step generative models, our approach makes sample-based approximation of DPO loss computationally feasible. Furthermore, the sample-based formulation makes FestDPO agnostic to the model family and sampling procedure. Our toy experiment demonstrates that FestDPO matches the reward-tilted target distribution across four few-step generators. For real-world tasks, we evaluate FestDPO in two domains: text-to-image generation and protein backbone generation. In text-to-image generation, FestDPO outperforms preference optimization baselines in both win rates against the base models and human evaluation scores. In protein backbone generation, it achieves a higher

β

-sheet fraction and better structural designability than the baselines.

Jaewoo Lee, Kyuil Sim, Hyeongyu Kang +3

KAIST · MongooseAI · Omelet

Generative Models Diffusion Alignment Preference Alignment Learning Reward Functions

Sep 28, 2026cs.LG

Tilted Schrödinger Bridge Matching

Schrödinger bridges provide an entropy-regularized framework and a principled solution for unpaired domain translation. In practice, a pretrained bridge may need to be adapted to human preferences or physical constraints through a reward a problem closely related to reward tilting in diffusion models but underexplored for Schrödinger bridges. We introduce Tilted Schrödinger Bridge Matching (TSBM), a post-training method for fine-tuning a learned bridge

P

between source

p_0

and target

p_1

toward a reward-tilted target

p_1^r\propto p_1e^r

, while preserving source

p_0

. We formulate this adaptation as alternating optimization initialized from

P

, provide theoretical justification, and derive a practical algorithm based on Adjoint Matching. We evaluate TSBM on unpaired image-to-image translation targeting digit properties in MNIST and facial attributes in CelebA.

Sergei Kholkin, Evgeny Burnaev, Alexander Korotin

Applied AI Institute

Schrödinger Bridges Unpaired Image-To-Image Translation Missing Bridge Bridge Diffusion Alignment Differentiable Optimal Transport+1

Sep 21, 2026cs.CV

AlignMorph: Tuning-Free Diffusion Image Morphing via Explicit Semantic Transport

Image morphing aims to produce a smooth and semantically consistent transition between two input images. Existing diffusion-based morphing methods either require expensive per-pair optimization or rely on implicit spatial alignment, which easily fails under large layout discrepancies. To address these limitations, we propose AlignMorph, a novel tuning-free diffusion framework guided by the principle of transport-then-denoise. We explicitly decouple geometric alignment from generative denoising to avoid structural entanglement. Our framework consists of two core components. (1) Global Semantic Transport, which achieves diffusion-compatible semantic alignment via entropic optimal transport and reliability-aware latent warping; and (2) Coordinate-Aligned Generation, which uses a symmetric bi-phase attention handoff to maintain consistent spatial coordinates throughout denoising. Without any tuning, AlignMorph effectively eliminates ghosting and achieves superior structural coherence and temporal smoothness on morphing benchmarks. Code is available at https://github.com/51xOne/Alignmorph.

Wuyi Liu, Xu Han, Yuren Chen +3

Huazhong University of Science and Technology · Beijing Jiaotong University · Beihang University +1

Diffusion Alignment Iterative Denoising Methods

Sep 16, 2026cs.CV

Beyond Random Couplings: Contrastive Noise Alignment in Generative Flows

Diffusion and flow-matching models are typically trained by corrupting data through independently sampled Gaussian noise. While simple and scalable, this forward process induces arbitrary data-noise couplings, forcing the network to learn high-curvature transports between unrelated endpoints. Existing optimal-transport methods reduce this burden by reassigning fixed noise samples to data, but the source noise distribution itself remains passive. To address this, we introduce Contrastive Noise Alignment (CNA), a training-time method that creates dynamic, contrastive couplings by optimizing the noise representations directly. By modeling the noise batch as an interacting particle system, CNA employs a cross-modal InfoNCE objective to align noise particles with their paired data targets. To prevent spatial collapse, this alignment is regularized using an angular entropy term and a radial norm penalty. We show theoretically that this equilibrium asymptotically preserves Gaussian structures, maintaining tractability during inference. Empirically, CNA improves the alignment between noise and data, reduces flow curvature, and provides better generation quality with fewer required sampling steps. For few-step, pixel-space generation (2-4 NFEs), CNA reduces FID by over 50% compared to standard rectified flow, and by at least 24% against Optimal Transport baselines.

Lennart Wittke, Vinicius Azevedo

ETH Zürich · Disney Research | Studios

Diffusion Alignment Generative Flow Networks Local Curvature

Sep 14, 2026cs.LG

DCRA: Diffusion-Conditioned Representation Alignment for Robust Time-Series Learning

Learning robust representations for time-series signals under noise and distribution shifts remains challenging, especially in clinical applications such as electroencephalogram (EEG) and electrocardiogram (ECG) analysis. We propose Diffusion-Conditioned Representation Alignment (DCRA), a training framework that repurposes the forward diffusion process as a structured corruption scheduler for representation learning. Different from conventional augmentation and consistency-based methods that rely on independently sampled perturbations, DCRA introduces a structured corruption trajectory via the diffusion forward process, which enables continuous and controlled representation evolution across noise levels. We introduce a feature-level consistency objective that aligns representations across noise levels while preserving class-discriminative structure. This mechanism promotes structure-preserving consistency, which enables smooth and semantically coherent feature trajectories in latent space. The proposed framework is encoder-agnostic and can be integrated with state space models and Transformer architectures. The seizure detection experiments on the CHB-MIT EEG dataset show that DCRA consistently improves performance under multiple noise conditions and achieves higher sensitivity at low false-positive rates. Analysis reveals that DCRA produces more balanced and structured representations compared to baseline and diffusion-only models. These findings highlight the benefit of combining structured corruption with representation alignment for robust time-series learning.

Wenrui Xu, Anas Enanaa, Keshab K. Parhi

University of Minnesota, Twin Cities

Seizure Detection Representation Learning Electrocardiogram Foundation Models Diffusion Alignment Adversarial Corruption

Sep 9, 2026stat.ML

FlowCPO: A Unified Divergence View of Preference Alignment for Flow Models

Preference alignment for flow and diffusion models now spans online reinforcement learning and offline preference optimization, but the relation between these methods remains unclear. In particular, existing forward-process alignment methods require fresh samples from the current model, while offline methods based on fixed preference pairs rely primarily on positive-only fine-tuning or DPO-style likelihood-ratio surrogates. We organize these approaches through a divergence-based framework and introduce FlowCPO, an offline forward-KL objective that uses both preferred and dispreferred samples without online rollouts. For linear interpolation, we show under explicit regularity conditions that the forward-KL objective is bounded by a contrastive flow matching loss, yielding a tractable surrogate on fixed data. We further show that this loss is nonnegative, whereas the signed regression loss of simplified FlowDPO can be unbounded below. In the in-domain setting, FlowCPO achieves higher mean GenEval and OCR scores than the evaluated baselines, reaching 0.84 and 0.87 versus 0.81 and 0.74 for FlowDPO at CFG 3.0. In the out-of-domain setting, the results are mixed, with the best GenEval result but lower reward scores than RFT on several metrics.

Yansen Han, Shengyi Liao, Peng Sun +4

Westlake University · Zhejiang University · Kling Team, Kuaishou Technology

Flow Models Preference Alignment Diffusion Alignment Kullback-Leibler Divergence

Aug 4, 2026cs.LG

Latent Reward Registers for Diffusion Preference Alignment

Aligning diffusion models with human preferences usually relies on a sparse terminal reward evaluated on the final generated samples, presenting a severe temporal credit-assignment challenge across the multi-step denoising process. We propose Latent Reward Registers, a mechanism that estimates terminal preference directly from intermediate noisy latents by prepending learnable, position-free register tokens to the input sequence of a frozen Diffusion Transformer (DiT). This independent readout mechanism extracts latent reward evidence without altering the generator's hidden states or velocity field. The resulting dense, differentiable reward signal throughout the full denoising process facilitates two alignment strategies. For training, Reward-Gradient On-Policy Distillation (RG-OPD) distills reward-guided updates along on-policy trajectories, bypassing the computationally expensive rollouts of standard policy gradients. For inference, Reward-Guided Sampling (RGS) steers trajectories via magnitude-matched reward gradients without parameter updates. Empirically, at high noise levels (u = 0.8), the registers reach the highest pairwise accuracy among the evaluated latent reward models. Furthermore, RG-OPD outperforms online reinforcement learning baselines while reducing GPU hours by up to 33x, and RGS establishes a new state-of-the-art among training-free methods, strictly enhancing both alignment and perceptual metrics. Code and weights are available at https://github.com/Guanys-dar/latent-reward-register

Yuanshen Guan, Zipeng Feng, Chengru Song +2

1Kling Team · University of Science and Technology of China

Diffusion Alignment Reward Gradients

Jul 30, 2026cs.CV

Temporal Concentration from Rollout Errors: Implicit Preference Optimization for Text-to-Video Diffusion

Recent advances in preference alignment for diffusion-based video generation, particularly via Direct Preference Optimization (DPO), have significantly improved visual quality. However, temporally sparse artifacts such as motion collapse, object flickering, and color oversaturation remain a major barrier to perceptual realism. Existing methods struggle with these issues due to two key limitations: (1) the preference attribution bottleneck, where offline human annotations are costly and fail to accurately capture learning dynamics, while online reward signals are rollout-aware but often unstable and biased; and (2) temporal credit misallocation, where uniformly applied supervision cannot effectively target the brief segments in which artifacts occur. To address these challenges, we propose concentrated Implicit Preference Optimization (cIPO), a post-training framework for video diffusion models. cIPO derives implicit preference signals directly from the denoising process: given a real video, the model adds forward noise and reconstructs it via iterative denoising, treating the original as the preferred sample and the reconstruction as the dispreferred one. This formulation captures inference-time errors without requiring human annotations or external reward models. Moreover, frame-level discrepancies between original and reconstructed videos reveal when failures occur. cIPO leverages this by computing temporal reconstruction errors and concentrating optimization on high-error segments, enabling more precise correction of failure-prone regions. Extensive experiments demonstrate that cIPO consistently enhances video authenticity and temporal coherence across multiple datasets, highlighting the effectiveness and efficiency of implicit preference with temporally concentrated optimization.

Henglin Liu, Fangyuan Kong, Jing Wang +7

Tsinghua University · Kling Team, Kuaishou Technology · Sun Yat-sen University

Text-To-Video Diffusion Models Video Diffusion Models Diffusion Alignment Sequential Preference Optimization Authenticity Rollout

Jul 29, 2026cs.CV

Anchoring and Steering Diffusion: Enhancing the Faithfulness of Text-to-Image Generation at Inference Time

While text-to-image diffusion models achieve impressive visual quality, they frequently struggle to maintain precise alignment with complex compositional prompts. An effective strategy is to improve the inference process of diffusion models, thereby better leveraging their pretrained priors to address misalignment. Existing training-free methods can be divided into two categories. The first category focuses on improving the randomly sampled initial noise, either performing costly search over noise pools or manipulating sampled noise without ensuring reliable semantic injection. The second category focuses on improving the denoising trajectory, lacking explicit mechanisms to timely diagnose and correct semantic errors. we propose \textbf{AnchorSteer}, a training-free framework that exerts fine-grained control over \textbf{both initialization} and \textbf{the denoising trajectory}. AnchorSteer consists of two synergistic components: \textbf{Semantic Anchoring} replaces uninformative Gaussian noise with text-aligned initializations via CLIP-based prior extraction and a novel Latent-Prior Score Distillation Sampling (LP-SDS) objective. Specifically, LP-SDS distills CLIP visual priors into the knowledge distribution of diffusion models, mitigating the domain gap between CLIP-based priors and diffusion-based priors. \textbf{Reflective Steering} transforms passive denoising with an active Think--Erase--Retouch loop that enables mid-generation self-correction. It leverages VLM-based diagnosis to detect semantic deviations and performs targeted latent refinement to suppress erroneous content and recover missing attributes. Extensive experiments on GenEval and T2I-CompBench++ demonstrate that AnchorSteer consistently outperforms existing baselines in text--image alignment while preserving high visual quality.

Xinyi Wang, Yuyang Huang, Yalin Su +4

Shanghai Jiao Tong University, Shanghai, China

Text-To-Image Diffusion Models Diffusion Alignment Contrastive Language-Image Pre-Training Model Semantic Anchor Dataset Distillation Steering

Jul 20, 2026cs.LG

DiFA: Inference-Time Forward-Process Alignment for Diffusion Models

The prevailing inference framework for diffusion models formulates generation fundamentally as a problem of numerical integration. This perspective casts the model as an exact estimator, neglecting the inherent statistical uncertainty of the denoising process. In this work, we propose Forward-Process Aligned Diffusion prediction (\textbf{DiFA}), a training-free framework that reframes inference-time data prediction refinement as a sequential state estimation problem. Rather than reusing past outputs solely for numerical integration, DiFA treats iterative data predictions along the reverse trajectory as correlated observations to build a forward-aligned temporal consensus. Inspired by Kalman filtering, this consensus aggregates historical predictions according to structural consistency and noise-level compatibility. To counteract the over-smoothing tendency of temporal consensus, we introduce a deviation guidance mechanism to adaptively preserve residual details. Empirically, DiFA yields significant improvements on CIFAR-10 and ImageNet across the evaluated metrics, including FID, IS, and FD-DINOv2, demonstrating that aligning inference with the forward statistical structure substantially improves generative fidelity.

Shigui Li, Delu Zeng

School of Mathematics, South China University of Technology, Guangzhou, China · School of Electronic and Information Engineering, South China University of Technology, Guangzhou, China

Diffusion Alignment Diffusion Models Distributional Alignment Inference-Time

Jul 19, 2026cs.CV

HarmoHOI: Harmonizing Appearance and 3D Motion for Multi-view Hand-Object Interaction Synthesis

Hand-Object Interaction (HOI) synthesis is a cornerstone for animation production and embodied AI. Despite the strong priors of video foundation models, multi-view consistent HOI synthesis remains challenging due to complex hand motions and occlusions. We present HarmoHOI, a unified diffusion framework that jointly and harmoniously generates synchronized multi-view HOI videos and globally aligned 3D point tracks. Our core insight is that robust multi-view consistency fundamentally requires globally aligned 3D geometry and motion. To this end, we propose a Mixture of Multi-view Diffusion Transformer that co-models RGB videos and 3D point tracks. By representing point tracks as pseudo-videos, we align 3D geometric signals with the 2D latent space of foundation models, thereby minimizing the domain gap and easing adaptation of priors. To further ensure geometry consistency, we introduce Global Motion Aligning Diffusion, which refines coarse point tracks into metric-scale, globally aligned 3D trajectories. HarmoHOI enables on-the-fly co-evolution of 2D appearance and 3D motion during denoising. To overcome the scarcity of multi-view HOI data, we employ a hybrid data curriculum learning strategy that successfully transfers generic priors from single-view data to synchronized multi-view generation. Experimental results show that HarmoHOI achieves state-of-the-art performance in visual quality, motion plausibility, and multi-view geometric consistency. Project page available at https://droliven.github.io/HarmoHOI_project.

Lingwei Dang, Juntong Li, Zonghan Li +5

South China University of Technology, Guangzhou, China · Beijing Normal University, Beijing, China · Tsinghua University, Beijing, China +1

Hand-Object Interaction Detection 3D Motion Video Foundation Models Diffusion Alignment Harms

Jul 15, 2026cs.LG

Integration Matters: Rollout-Based Training for Constrained Diffusion Models

Constrained generative models aim to produce samples that satisfy complex feasibility constraints while remaining faithful to the data distribution. Existing constrained generation methods typically enforce constraints either through training-time optimization or sampling-time correction. Training-time optimization approaches optimize on states induced by the training distribution, which can differ substantially from those encountered during sampling. Sampling-time correction methods instead modify the sampling process at inference, introducing distribution shift and requiring expensive tuning, particularly for few-step sampling. We propose a fine-tuning framework that incorporates constraint guidance obtained through online rollout into the training process, which aligns training with sampling by differentiating through the fixed noise schedule used to numerically integrate the denoising process. This exposes the model to violations that arise along the denoising trajectory and aligns diffusion learning with the sampling process. Experiments across multiple tasks show that our method improves constraint satisfaction while maintaining competitive sampling quality compared to prior methods.

Xiaoxuan Liang, Saeid Naderiparizi, Berend Zwartsenberg +1

University of British Columbia · Inverted AI · Alberta Machine Intelligence Institute

Diffusion Models Diffusion Alignment Generative Models Training Time

Jul 3, 2026cs.AI

Efficient bias mitigation in T2I diffusion models using Concept Graphs

Text-to-Image diffusion models often propagate harmful bias inherited from the training data. Existing bias mitigation techniques typically intervene only at the text encoder or provide inference-time guidance, often leading to generations that collapse into semantically incoherent outputs. To address these limitations, we introduce CO-ALIGN (Concept Ontology Alignment), a novel bias mitigation approach based on concept-graph alignment that operates on the model's internal concept ontology. By aligning concepts within the text encoder and denoiser, CO-ALIGN achieves substantial bias reduction while preserving generative integrity. We demonstrate the effectiveness of concept-graph alignment across three paradigms: text-encoders, denoisers and joint text-denoiser ontology alignment. CO-ALIGN outperforms the state of the art, improving fairness by

30\%

,

ΔFID=11.4

in image quality,

2.8\%

in image fidelity, all while reducing semantically incoherent outputs by

88\%

. Beyond bias mitigation, we show that CO-ALIGN benefits other downstream tasks as well. In particular, our experiments demonstrate that better-aligned internal ontologies enhance concept unlearning robustness across multiple unlearning techniques.

Mansi, Avinash Kori, Francesco Leofante

Department of Computing Imperial College London

Text-To-Image Diffusion Models Diffusion Alignment Image-Text Alignment Diffusion Models Hierarchical Semantic Alignment Unlearning Method

Jul 2, 2026cs.CV

Open-Weather Robust 3D Detection via Dual-Critic Diffusion Alignment

Robust 3D object detection under adverse weather remains a critical hurdle for autonomous driving. Despite progress with LiDAR-4D radar fusion, most methods are constrained by a closed-world assumption, implicitly requiring training and test weather to align in both type and severity. This premise fails in practice: the open-ended nature of weather, and even variations within a single type like rain, cause dramatically different LiDAR degradation patterns, leading to significant performance drops in unseen conditions. To address this, we present Dual-Critic Guided Diffusion Alignment (DCDA), a weather-agnostic framework that learns to recover degraded LiDAR features toward a clean manifold. Rather than modeling specific weather types, DCDA employs a 4D radar-conditioned diffusion process to progressively refine features, guided by two complementary critics. (i) A detection-guided critic, anchored by a pre-trained clean-weather model, ensures that the refined features retain object-level discriminability and localization accuracy. (ii) A weather adversarial critic enforces holistic distributional consistency with clean-weather representations. By aligning features through semantic and distributional constraints rather than explicit weather modeling, DCDA generalizes effectively to unseen weather types and severities without requiring paired data or weather labels. We further introduce a structured open-weather benchmark with held-out type-severity combinations and extensive experiments verify DCDA's advantages.

Shuyao Li, Chuanxing Geng, Heyang Sun +2

Nanjing University of Aeronautics and Astronautics, Nanjing, China · MIIT Key Laboratory of Pattern Analysis and Machine Intelligence, Nanjing, China

3D Object Detection Autonomous Driving Perception Diffusion Alignment

Jul 1, 2026cs.LG

PAPA: Online Personalized Active Preference Alignment

Diffusion models are highly effective at modeling complex data distributions, including images and text. However, in applications like personalized recommender systems, the objective often shifts to modeling specific regions of the distribution that maximize user preferences-initially unknown but gradually uncovered through interactive feedback. This can naturally be framed as a reinforcement learning problem, where the goal is to fine-tune a diffusion model to maximize a reward function based on preferences. However, the main challenge lies in learning a parameterized reward model, which typically requires large-scale preference data-something that is often not feasible in practice. In this work, we introduce Personalized Active Preference Alignment PAPA, a novel method that bypasses the requirement for a parametrized reward model by directly optimizing the diffusion model using real-time user feedback. PAPA enables feedback-efficient preference alignment, drawing inspiration from the variational inference framework. We demonstrate PAPA's effectiveness through extensive experiments and ablation studies across diverse class-conditioned and fine-grained alignment tasks. Additionally, based on theoretical insights, we propose an enhanced fine-tuning strategy, referred to as EPAPA, that requires less computational budget and accelerates the fine-tuning process, further boosting PAPA's suitability for real-world deployment. Our code is made publicly available at https://github.com/NasikNafi/papa.

Anindya Sarkar, Nasik Muhammad Nafi, Isaac Lyngaas +2

Washington University in St. Louis · Oak Ridge National Laboratory

Preference Alignment Diffusion Alignment Diffusion Models Reward Functions Model Fine-Tuning

Jul 1, 2026cs.CV

The Illusion of High Utility in Safety Alignment of Text-to-Image Diffusion Models

Safety alignment of text-to-image (T2I) diffusion models aims to suppress harmful generations while preserving utility on benign prompts. Recent methods often appear to deliver high safety with high utility, but this conclusion rests largely on coarse global utility metrics (e.g., FID, CLIPScore) that are insensitive to fine-grained semantic correctness, creating an illusion of high utility. We show that when utility is measured with structured evaluation, this illusion breaks: on TIFA (Text-to-Image Faithfulness evaluation with Question Answering), safety-aligned models suffer substantial drops in semantic fidelity, including failures in object counts, attributes, and relationships. To diagnose the source of this gap, we analyze the text-encoder prompt embedding space and uncover semantic collapse, a contraction of embedding spread coupled with distortion of inter-prompt similarity structure, which strongly correlates with structured utility loss. Guided by this insight, we propose StructureAware Geometric Regularization (SAGE), a safety alignment objective that explicitly preserves embedding spread and inter-prompt relational structure during adaptation. Our method restores structured utility (TIFA +5.0% over prior state-of-the-art) while maintaining strong safety performance and competitive coarse-grained utility scores. Our source code and trained models are available at https://adeelyousaf.github.io/SAGE_ECCV26_Project_Page/.

Adeel Yousaf, Soumik Ghosh, James Beetham +2

Institute of Artificial Intelligence University of Central Florida, Orlando, United States

Text-To-Image Diffusion Models Text-To-Image Image-Text Alignment Diffusion Alignment Safety Alignment Diffusion Models+2

Jun 28, 2026cs.LG

Beyond Trajectory Matching: Reflow with Marginal Distribution Alignment

Diffusion and continuous-flow generative models achieve high-quality generation, and their deterministic sampling can be formulated as solving learned ODE dynamics. However, accurate ODE discretization often requires many steps, making efficient few-step generation a key challenge. Among acceleration strategies, reflow-based distillation simplifies teacher ODE trajectories so that a student model can approximate the teacher transport with fewer steps. We identify a theoretical limitation of this paradigm, namely that trajectory matching can under-determine the distribution induced by the student model. In particular, two student models can attain the same trajectory-matching loss while inducing different endpoint marginal distributions, which may lead to different generation quality. To address this limitation, we introduce a marginal-alignment regularizer that penalizes the discrepancy between the student-induced marginal and the corresponding teacher marginal at the endpoint of each distillation interval. The regularizer is computed by tracking log-density changes along the ODE induced by the student model and evaluating scores from the frozen teacher model, without requiring auxiliary trainable networks or adversarial optimization. The resulting framework applies uniformly to the reflow family, including vanilla reflow and piecewise reflow. We further prove a telescoping total-variation bound showing that local marginal alignment controls the final-time discrepancy between the student-induced and teacher-induced distributions. Experiments on benchmark backbones demonstrate the effectiveness of the proposed method for few-step generation.

Chen Wang, Peiran Yun, Pan Xie +1

Department of Statistics and Data Science, Tsinghua University · Department of Computer Science and Technology, Tsinghua University · Bytedance Inc

Diffusion Alignment Distribution Matching Distillation Generative Models Distributions Time Discretization Trajectory Generation+1

Jun 23, 2026cs.CV

FLUX3D: High-Fidelity 3D Gaussian Generation with Diffusion-Aligned Sparse Representation

Sparse voxel representation has emerged as a scalable foundation for image-to-3D Gaussian Splatting (3DGS) generation, yet current methods struggle to preserve high-frequency visual details of input images due to two structural bottlenecks. First, they adopt discriminative 2D features optimized for semantic abstraction to construct sparse voxel latents, which suppress reconstructive cues and induce a representation bottleneck. Second, in the generation stage, standard diffusion transformers lack effective mechanisms to align dense 2D image tokens with sparse 3D voxel latents, resulting in a cross-modal correspondence bottleneck. To address these issues, we propose FLUX3D, a scalable image-to-3DGS framework that boosts both representation learning and cross-modal alignment during generation. We first revisit 2D feature selection for sparse-voxel-based 3D representation learning, propose Diffusion-Aligned Structured Latents (DA-SLAT) and couple it with a decoder-only architecture to improve 3DGS reconstruction fidelity. We also design a sparse-structure-aware diffusion framework, which integrates the Sparse-structure Multimodal Diffusion Transformer (SMDiT) and Modal-Aware Rotary Positional Embedding (MARoPE) to achieve geometry-agnostic 2D-3D alignment. Extensive benchmark experiments demonstrate that FLUX3D yields substantial improvements in appearance fidelity and significantly outperforms all state-of-the-art (SOTA) methods in generating high-quality 3DGS assets.

Haorui Ji, Weizhe Liu, Hongdong Li +1

The Australian National University · ByteDance

3D Gaussian 3D Generative Models Voxel 3D Generation Diffusion Alignment Diffusion Transformers+1

Jun 22, 2026cs.LG

PG-MAP: Joint MAP Optimization for Inference-Time Alignment of Diffusion and Flow-Matching Models

Inference-time alignment of pretrained text-to-image models is typically performed along a single control axis, such as classifier-free guidance, attention editing, or reward-based latent perturbations. This limitation prevents modeling joint dependencies between conditioning and latent variables and hinders transfer across generative transports. We propose PG-MAP, a training-free framework that formulates inference-time alignment as a trajectory-level Gibbs-MAP / proximal energy optimization over the conditioning

c

and latent state

z_t

via a forward-consistency coupling, optionally guided by a frozen preference reward. This joint formulation enables coordinated updates across modalities while remaining compatible with both diffusion and flow-matching models through transport-specific adaptations. Across diffusion backbones (SD~1.5, SDXL), PG-MAP consistently improves alignment metrics such as PickScore and Aesthetic, and can be effectively combined with tuned classifier-free guidance to achieve the strongest overall performance. On flow-matching models (SD3.5-medium), the framework reduces to a latent-only variant, achieving

\mathbf{91.9\%}

PickScore and

75.7\%

HPS win rates against a static baseline, with controlled experiments ruling out noise-related artifacts. Human evaluations further confirm consistent preference over strong baselines, including tuned CFG and compute-matched universal guidance. Finally, an oracle-routing analysis shows that the relative importance of conditioning and latent optimization depends on prompt types, surfacing further headroom that a per-prompt selector could exploit.

Ruolan Sun, Pawel Polak

Stony Brook University

Diffusion Alignment Contrastive Flow Matching Mean Average Precision Text-To-Image

Jun 17, 2026cs.LG

The Reward Was in Your Data All Along: Correcting Flow Matching with Discriminator-Guided RL

Score- and flow-matching models often rely on preference-based reinforcement learning for two purposes: aligning with subjective preferences and, surprisingly, recovering properties such as visual realism and coherent object structure that matching-based training is intended to learn from the data itself. We argue that this reflects a structural mismatch. Matching losses measure

\ell_2

regression error on the velocity or score field under training-time marginals, a proxy poorly aligned with the visual and semantic properties that determine sample quality at inference. Given a reward aligned with these properties, RL sidesteps the mismatch by evaluating the model on its own samples and following the reward landscape directly. The challenge is to obtain such a reward without relying on human preferences, which are expensive and conflate data realism with annotator inclinations. We propose Discriminator-Guided RL (DRL). DRL trains a discriminator to separate data from base-model samples in a pretrained representation space and uses its logit as the reward in KL-regularized RL. The pretrained space restricts the discriminator to perceptually meaningful directions, and the logit estimates the log-likelihood ratio between data and model, which is the optimal reward for targeting the data distribution. Across SiT, JiT, REPA, and RAE, DRL reduces guidance-free FID (e.g.,

9.38 \to 2.62

on SiT) and semantic-space FD (e.g.,

88.2 \to 19.3

on DINOv3 for SiT), with consistent gains across all backbones, and improves human-preference rewards without training on them. It also yields a better Pareto frontier between preference reward and image fidelity under subsequent preference-based post-training, increasing alignment while reducing low-level artifacts such as oversaturation and excessive brightness.

Nicolas Beltran-Velez, Felix Friedrich, Zhang Xiaofeng +4

FAIR at Meta · Columbia University · Mila – Québec AI Institute +3

Contrastive Flow Matching Diffusion Alignment Post-Training

Jun 7, 2026cs.CV

MaskAlign: Token-Subset Representation Alignment for Efficient Diffusion Training

Representation alignment with pretrained vision models has recently shown strong potential for accelerating diffusion transformer training. By aligning intermediate diffusion features with clean-image representations from self-supervised vision encoders, existing methods improve convergence and generation quality. However, such alignment also introduces a non-trivial constraint: diffusion models operate on noisy inputs whose usable information varies across timesteps, while the reference features are extracted from clean images. In this paper, we revisit this mismatch from a token-level perspective. We find that, under full-token representation alignment, tokens with large alignment-gradient norms exhibit a stable spatial preference, suggesting that the alignment objective does not affect all tokens uniformly and may encourage the model to rely on the complete set of clean-image tokens. To address this issue, we propose MaskAlign, a token-subset representation alignment method that applies alignment to randomly sampled token subsets during training. By exposing the model to different token subsets across iterations, MaskAlign reduces the dependence of representation alignment on the complete token set and encourages alignment behavior that is more stable under token-subset perturbations. To mitigate the information loss caused by directly dropping tokens, we further introduce a lightweight pre-mask token mixing block that shares information across tokens before masking.

Lianyu Pang, Tianlin Pan, Cheng Da +5

The Hong Kong University of Science and Technology · Kuaishou Technology · University of Chinese Academy of Sciences

Diffusion Alignment Diffusion Transformers Vision Encoders

Jun 4, 2026cs.CV

Where, What, Why, and Importance: Structured Defect Grounding for Text-to-Image Feedback

Despite generating increasingly photorealistic images, text-to-image (T2I) models still exhibit localized, subtle, and structurally complex failures. Diagnosing these failures requires instance-level feedback that answers where a defect occurs, what type it is, why it is defective, and its importance to overall image quality. While recent dense-feedback methods move beyond scalar supervision, their heatmap-centric representations still formulate diagnosis as pixel-field regression, making it difficult to localize variable-cardinality defects and bind semantic reasons to individual failures. To address this representation bottleneck, we propose Structured Defect Grounding (SDG), which casts T2I diagnosis as structured set prediction by modeling each defect as a (location, type, reason, importance) tuple. To make this formulation trainable and measurable, we introduce SDG-30K, a 30K-image dataset with box-grounded annotations across four modern T2I generators, together with a dedicated evaluation protocol, SDG-Eval. Building on this structured representation, we further present a diagnosis-to-alignment framework in which a Vision-Language Model (VLM) serves as the SDG detector, and BoxFlow-GRPO converts predicted defect sets into box-derived, importance-weighted spatial rewards for diffusion model alignment. Extensive experiments show that our SDG detector outperforms leading proprietary VLMs on structured defect grounding, while SDG-guided rewards consistently improve T2I alignment and support localized image refinement. These results establish SDG as a unified, instance-level interface for diagnosing, evaluating, and enhancing modern generative models.

Huaisong Zhang, Hao Yu, Yuxuan Zhang +7

Tsinghua University · University of British Columbia · Vector Institute +2

Text-To-Image Diffusion Alignment Defect Contextual Grounding Feedback

Jun 1, 2026cs.LG

Drifting Preference Optimization for One-Step Generative Models

One-step text-to-image generators are attractive for deployment because they generate an image with a single forward pass, but preference finetuning them remains difficult: standard alignment methods often rely on policy likelihoods, denoising trajectories, differentiable reward gradients, or test-time optimization. We propose Drifting Preference Optimization (DrPO), an online preference-finetuning method for deterministic one-step generators. For each prompt, DrPO samples candidates from the current generator, ranks them with a target reward, and uses high- and low-scoring samples to synthesize a feature-space update direction. The update is a non-parametric dipole preference field plus a reference drift estimated from the frozen base generator, and is optimized through a detached feature-space regression target. The target reward is used only for ranking, so DrPO can train with large, black-box, or non-differentiable rewards while inference remains a single generator call. We evaluate DrPO on SD-Turbo and SDXL-Turbo with multiple target rewards and benchmarks, including HPSv3 and GenEval. DrPO improves alignment over reward-gradient-free one-step preference baselines and reduces HPSv3 training computation by

3.51\times

under the matched effective-batch setting by removing reward-model backpropagation. Initial offline experiments suggest that sample-based gradient synthesis can also be used beyond online reward ranking.

Zhou Jiang, Yandong Wen, Zhen Liu

Westlake University · The Chinese University of Hong Kong, Shenzhen

Diffusion Alignment Few-Step Generation Generative Models Preference Alignment Learning Backpropagation

Jun 1, 2026cs.CV

FairGen: Preference-Aligned Diffusion for Demographically Equitable Medical Image Synthesis

Medical imaging is central to modern diagnostics, and artificial intelligence (AI) systems are increasingly used to support image-based analysis by improving efficiency, accuracy, and access to care. However, inequities in healthcare access and differential disease prevalence create severe demographic imbalances in clinical image data. Such imbalances are compounded by the fact that diseases can manifest with distinct features across demographic groups, rendering certain phenotypic presentations naturally rare. AI models trained on such imbalanced data risk perpetuating diagnostic bias and widening healthcare disparities. Here we introduce FairGen, a fairness-aware diffusion framework that synthesizes demographically balanced medical images while preserving pathology-relevant visual features. By embedding physician-aligned preferences into the generation process, FairGen improves subgroup coverage during synthesis and downstream classification. Applied to dermatology, radiology, and neuroimaging benchmark tasks, FairGen achieves fairness improvements of 95.9% for skin images, 80.0% for chest radiography, and 35.2% for brain MRI, while maintaining competitive diagnostic accuracy relative to models trained on original clinical data. Clinician-facing expert review and external validation on independent cohorts further support that these gains extend beyond standard fidelity metrics and are not confined to the original in-distribution datasets.

Zhimin Li, Ruichen Zhang, Zhen Tan +3

University of Pittsburgh, Swanson School of Engineering, Pittsburgh, 15261, USA · 2The University of North Carolina at Chapel Hill, Department of Computer Science, Chapel Hill, 27599, USA · 3Arizona State University, School of Computing and Augmented Intelligence, Tempe, 85281, USA +1

Medical Image Generation Diffusion Alignment Diffusion Models Diagnosis Synthesis

May 30, 2026cs.CV

Improving Visual Representation Alignment Generation with GRPO

Recent diffusion transformers have demonstrated strong image synthesis capabilities but remain inefficient to train due to weak alignment between generative and discriminative representations. While representation alignment frameworks such as REPA improve convergence by aligning noisy denoising features with pretrained visual encoders, their externally supervised alignment loss is static and lacks adaptivity during training and inference. Existing methods rely on fixed cosine alignment or contrastive objectives, which cannot dynamically balance representation consistency and generation quality, resulting in limited discriminative benefit and failing to optimize alignment in a task-adaptive manner. To address this, we propose VRPO, a reinforcement-based optimization strategy that replaces REPA's static alignment loss with a generative representation policy optimization objective. Instead of enforcing a fixed similarity constraint, VRPO treats representation alignment as a reward-guided process: the model receives adaptive rewards based on generation fidelity, perceptual quality, and semantic coherence between the diffusion features and pretrained visual embeddings. This formulation enables the generator to continuously refine its internal representations toward semantically meaningful directions while improving image quality. Our VRPO-driven training seamlessly integrates into diffusion transformers, introducing negligible computation cost and preserving full compatibility with SiT and DiT architectures. Extensive experiments on ImageNet-256x256 demonstrate that our VRPO-Alignment substantially enhances both convergence and fidelity, achieving up to +1.8 FID improvement and 2.3x faster training compared to REPA under identical compute budgets.

Shentong Mo, Sukmin Yun

Carnegie Mellon University · Hanyang University

Diffusion Alignment Diffusion Transformers Vision Encoders

May 28, 2026cs.LG

Alignment-Guided Score Matching for Text-to-Image Alignment in Diffusion Models

Diffusion models generate highly realistic images but often struggle with precise text-image alignment. While recent post-training methods improve alignment using external rewards or human preference signals, their performance heavily depends on reward quality and does not directly address alignment within the diffusion process itself. Recent reward-free approaches such as SoftREPA demonstrate that optimizing soft text tokens via contrastive learning can effectively improve text-image representation alignment, outperforming standard parameter-efficient fine-tuning baselines. However, the contrastive formulation can excessively penalize negative pairs, which manifests as characteristic failure cases such as over-counting and repetition. To address this issue, we propose a lightweight, reward-free post-training method that refines soft tokens by integrating contrastive alignment guidance directly into the score-matching objective of diffusion models. By assigning alignment directions at the score level, our approach mitigates these limitations and yields more coherent and semantically faithful generations. Experiments show that our method matches SoftREPA while substantially improving its failure cases, achieving over 35% improvement in counting accuracy on the GenEval benchmark. Our method is seamlessly applicable to existing diffusion backbones (SD1.5, SDXL, and SD3), and is complementary to existing RL-based diffusion post-training methods. Project page: https://jaayeon.github.io/AGSM

Jaa-Yeon Lee, Yeobin Hong, Taesung Kwon +1

Graduate School of AI, KAIST, South Korea.

Text-To-Image Diffusion Models Diffusion Alignment Image-Text Alignment Diffusion Models Contrastive Alignment Post-Training

May 26, 2026cs.LG

Explicit Critic Guidance for Aligning Diffusion Models

Online reinforcement learning is becoming increasingly important for aligning diffusion models with non-differentiable objectives. However, existing methods still face limitations in assigning fine-grained credit along denoising trajectories and in realizing stable value-based optimization. We propose a state-aligned latent actor-critic framework for diffusion post-training, in which the diffusion model serves as its own timestep-conditioned value function and predicts values directly on noisy latent states. This enables trajectory-level PPO training, supports stable actor-critic optimization with simple conditioning and value pretraining strategies, and naturally allows the learned critic to be reused for inference-time steering. We further extend the framework to multi-reward optimization, where joint training with complementary rewards helps alleviate reward hacking. Across both UNet- and DiT-based backbones, our method consistently outperforms prior group-relative RL and actor-critic baselines on single-reward and multi-reward benchmarks, while test-time steering provides additional gains in generation quality.

Zhengyang Liang, Qihang Zhang, Ceyuan Yang

University of Toronto · Vector Institute · The Chinese University of Hong Kong

Diffusion Alignment Diffusion-Based Reinforcement Learning Methods Diffusion Models Critic-Free Reinforcement Learning Soft Actor-Critic Offline Reinforcement Learning+1

May 26, 2026cs.LG

Beyond Pairwise Preferences: Listwise Reward-Aware Alignment for Diffusion Models

Preference optimization has emerged as an efficient alternative to online reinforcement learning from human feedback (RLHF) for aligning text-to-image diffusion models. However, existing methods largely reduce supervision to binary pairwise comparisons. This pairwise reduction is limiting when training data naturally contains multiple candidate images for the same prompt, and when continuous reward scores can provide richer information than a single winner-loser label. To address these limitations, we propose Diffusion LAIR, a reward-aware listwise preference optimization method for diffusion models. For each prompt, LAIR converts reward scores across a group of candidate images into centered advantage weights, then optimizes an advantage-weighted regression objective on the implicit reward, defined as the denoising-loss improvement of the current model over a fixed reference model, with a quadratic penalty that regularizes the magnitude of the implicit reward. The resulting objective uses all candidates simultaneously rather than selecting pairs, and remains conservative by explicitly controlling the magnitude of the implicit reward. The LAIR objective admits a bounded closed-form optimum in implicit-reward space, clarifying how the regularization strength controls the magnitude of the preference update. Experiments show that Diffusion LAIR outperforms strong preference optimization baselines on SD1.5 and SDXL across text-to-image generation, compositional generation, and image editing benchmarks.

Austin Wang, Jiaqi Han, Stefano Ermon +1

Caltech · Stanford University

Diffusion Alignment Text-To-Image Diffusion Models Backtranslation Augmented Direct Preference Optimization Diffusion Models Preference Alignment Learning

May 25, 2026cs.CV

DRM: Diffusion-based Reward Model With Step-wise Guidance

Current mainstream methods of aligning diffusion models with human preferences typically employ VLM-based reward models. However, these reward models, pre-trained for semantic alignment, struggle to capture the essential perceptual qualities-such as aesthetics, composition, and visual harmony. In this work, we argue that a model capable of high-fidelity generation must possess a profound understanding of these visual attributes. Based on this insight, we introduce the Diffusion-based Reward Model (DRM), a novel paradigm that use the pre-trained diffusion model as a powerful evaluative backbone. A key advantage of the DRM is its unique ability to assess not only the final image but also the noisy intermediate latents at any stage of the generative process. We leverage this step-wise evaluative capacity in two ways. First, we propose Step-wise GRPO, a reinforcement learning algorithm that provides dense, per-step rewards to resolve the imprecise credit assignment problem in GRPO algorithm, leading to more stable and effective alignment. Second, we introduce Step-wise Sampling, a novel inference strategy that employs the DRM as a dynamic guide to evaluate multiple generation paths at each step, steering the process towards higher-quality outcomes. Extensive experiments confirm that our approach significantly enhances the final quality of generated images. Code: https://github.com/jjaxonx/DRM.

Jaxon Zhang, Binxin Yang, Hubery Yin +2

Peking University · Work done during an internship at WeChat Vision, Tencent Inc. · WeChat Vision, Tencent Inc.

Diffusion Alignment Diffusion Models

May 24, 2026cs.LG

Inference-Time Alignment of Diffusion Models via Trust-Region Iterative Twisted Sequential Monte Carlo

We study inference-time alignment for diffusion-based generative models, aiming to steer a base model toward high-reward outputs without updating its weights. Recent Sequential Monte Carlo (SMC)-based steering methods approximate reward-tilted target distributions in a principled way, but their proposals remain largely tied to the base sampler. Since reward information is mainly used after propagation through particle reweighting and resampling, these methods can require large particle budgets and suffer from weight degeneracy and high-variance estimates. One way to reduce variance and improve particle efficiency is to iteratively learn twisting functions that provide look-ahead guidance, as in twisted SMC. However, existing learnable twisting methods are developed mainly for classical sequential inference and can be unstable when applied to diffusion-based alignment with high-dimensional state spaces and terminal, noisy, or black-box rewards. We propose Trust-Region Iterative Twisted Sequential Monte Carlo (TRI-TSMC), a trust-region framework for learning twisting functions in SMC-based inference-time alignment. Each iteration computes an exact KL-constrained update in path space, which admits a closed-form solution by tempered importance reweighting, and projects this target back to the parameterized twisted family by weighted maximum likelihood. Theoretically, we formalize the value-function interpretation of the optimal twisting function and show that it yields a zero-variance sampler. We prove that the trust-region update follows an escort path toward the target distribution, that the weighted maximum-likelihood update is a forward-KL projection, and that the path reduces residual importance-weight variance. Empirically, TRI-TSMC improves primary alignment objectives on discrete diffusion text generation and text-to-image generation under matched inference-time budgets.

Weixin Wang, Yu Yang, Wei Deng +1

Duke University · Morgan Stanley

Diffusion Alignment Diffusion Models Monte Carlo Latent Representation Alignment

May 19, 2026cs.CV

When Preference Labels Fall Short: Aligning Diffusion Models from Real Data

Preference alignment aims to guide generative models by learning from comparisons between preferred and non-preferred samples. In practice, most existing approaches rely on preference pairs constructed from model-generated images. Such supervision is inherently relative and can be ambiguous when both samples exhibit artifacts or limited visual quality, making it difficult to infer what constitutes a truly desirable output. In this work, we investigate whether real data can serve as an alternative source of supervision for preference alignment. We adopt a data-centric perspective and study a curation strategy that treats real images as reference points and constructs preference signals by contrasting them with generated or perturbed samples, without requiring manually annotated preference pairs. Through empirical analysis, we show that real-data-based supervision provides effective guidance for aligning diffusion models and achieves performance comparable to existing preference-based methods. Our results suggest that real data offers a practical and complementary source of supervision for preference alignment and highlight directions of label-efficient alignment strategies. Code and models are available at https://cwyxx.github.io/RealAlign.

Weiyan Chen, Weijian Deng, Yao Xiao +5

School of Computer Science and Engineering, Sun Yat-sen University, Guangzhou, China · Tsinghua Shenzhen International Graduate School, Tsinghua University, Shenzhen, China · Australian National University, Canberra, Australia +2

Diffusion Alignment Preference Alignment Diffusion Models Preference Alignment Learning Generative Models Short

May 19, 2026cs.CV

Stitched Value Model for Diffusion Alignment

For practical use, diffusion- or flow-based generative models must be aligned with task-specific rewards, such as prompt fidelity or aesthetic preference. That alignment is challenging because the reward is defined for clean output images, but the alignment procedure requires value function estimates at noisy intermediate latents. Existing methods resort to Tweedie-style or Monte Carlo approximations, trading off estimator bias against computational cost: Tweedie estimates are efficient but biased, while Monte Carlo estimates are more accurate but require expensive rollouts. A natural alternative would be a learned value function, but it remains an open question how to effectively train a strong and general value model specifically for noisy latents. Here, we propose StitchVM, a model stitching framework that efficiently transfers reward models pretrained for clean images to the noisy latent regime. StitchVM starts from an existing, truncated pixel-space reward model and attaches a frozen diffusion backbone to it as its head. From the pixel-space model, the resulting hybrid retains a carefully pretrained, robust reward capability; from the diffusion backbone, it inherits its native ability to handle noisy latents. The stitching procedure is exceptionally lightweight, e.g., stitching and finetuning CLIP ViT-L and SD 3.5 Medium takes only 10 GPU-hours. By lifting powerful pixel-space reward models to latent space, StitchVM opens up a new style of diffusion alignment: instead of rough, yet costly per-sample approximation of the value function, the correct function for the actual, noisy latents is constructed once and then amortized over many samples and iterations. We show that this approach yields improvements across a broad range of downstream steering and post-training methods: DPS becomes

3.2\times

faster while halving peak GPU memory, and DiffusionNFT becomes

2.3\times

faster.

Hyojun Go, Hyungjin Chung, Prune Truong +8

1ETH Zurich · 2Google · University of Copenhagen

Diffusion Alignment Diffusion Models Generative Flow Networks Sewing Value Functions Value

May 16, 2026cs.CV

Beyond Point-Wise Matching: Structural Representation Alignment for Accelerating Diffusion Transformers

Recent advances in Diffusion Transformers (DiTs) demonstrate that aligning noisy latent states with well-trained semantic features-as pioneered by Representation Alignment (REPA)-can substantially accelerate training and improve generation fidelity. Subsequent analysis(e.g., iREPA) suggests that these gains arise primarily from transferring spatial structure contained in pre-trained vision representations. However, mostly existing alignment methods employ point-wise matching objectives or rely on implicit architectural tweaks, which fail to explicitly model the spatial relational geometry inherent in vision foundation models. We argue that such element-wise supervision is insufficient to capture the rich spatial topology of visual representations, and that effective alignment for generation should instead be formulated as an explicit structural constraint. To this end, we propose sREPA, a structural REPresentation Alignment framework to enforce consistency in the relational geometry of feature maps, rather than merely matching individual feature points. By encouraging the model to internalize holistic spatial layouts and structural correlations from pre-trained features, sREPA achieves faster and more stable convergence, along with improved sample quality, compared to state-of-the-art alignment strategies. Our code and models will be released.

Shaodong Xu, Zhendong Wang, Litong Gong +4

University of Science and Technology of China

Diffusion Alignment Diffusion Transformers Harder Better Faster Denser Feature Matching Topology

May 15, 2026cs.CV

Flash-GRPO: Efficient Alignment for Video Diffusion via One-Step Policy Optimization

Group Relative Policy Optimization has emerged as essential for aligning video diffusion models with human preferences, but faces a critical computational bottleneck: training a 14B parametered model typically demands hundreds of GPU days per experiment. Existing efficiency methods reduce costs through sliding window subsampling training timesteps, but fundamentally compromise optimization, exhibiting severe instability and failing to reach full trajectory performance. We present Flash-GRPO, a single-step training framework that outperforms full trajectory training in alignment quality under low computational budgets while substantially improving training efficiency. Flash-GRPO addresses two critical challenges: iso-temporal grouping eliminates timestep-confounded variance by enforcing prompt-wise temporal consistency, decoupling policy performance from timestep difficulty; temporal gradient rectification neutralizes the time-dependent scaling factor that causes vastly inconsistent gradient magnitudes across timesteps. Experiments on 1.3B to 14B parameter models validate Flash-GRPO's effectiveness, demonstrating substantial training acceleration with consistent stability and state-of-the-art alignment quality.

Xiaoxuan He, Siming Fu, Zeyue Xue +9

Zhejiang University · Joy Future Academy · Independent Researcher +1

Video Diffusion Models Diffusion Alignment Flow-Grpo Value Alignment Subsampling

May 12, 2026cs.CV

CRAFT: Clinical Reward-Aligned Finetuning for Medical Image Synthesis

Foundation diffusion models can generate photorealistic natural images, but adapting them to medical imaging remains challenging. In medical adaptation, limited labeled data can exacerbate hallucination-like and clinically implausible synthesis, while existing metrics such as FID or Inception Score do not quantify per-image alignment with pathology-relevant criteria. We introduce the Clinical Alignment Score (CAS), a foundation-model-based proxy for clinical alignment that evaluates generated images along four complementary dimensions beyond visual fidelity. Building on CAS, we propose Clinical Reward-Aligned Finetuning (CRAFT), a reward-based adaptation framework that transfers medical knowledge from multimodal large language models and vision-language models through label-conditioned prompt enrichment, clinical checklists, and differentiable reward optimization. Across four diverse modalities, CRAFT improves CAS and downstream classification performance over strong adaptation baselines. Beyond average CAS gains, CRAFT reduces the empirical low-alignment tail below a real-image reference threshold by 5.5-34.7% points relative to the strongest baseline, corresponding to a 20.4% average relative reduction across datasets. These results indicate fewer hallucination-like generations under CAS, and are corroborated by out-of-family evaluator evaluation, structured checklist auditing, memorization analysis, and a blinded physician preference study on CheXpert.

Yunsung Chung, Alex El Darzi, Carlo El Khoury +3

Department of Computer Science, Tulane University, New Orleans, LA 70118 · School of Medicine, Tulane University, New Orleans, LA 70112

Medical Image Generation Diffusion Alignment Clinical Model Fine-Tuning Synthesis Reinforcement Fine-Tuning+1

May 12, 2026cs.LG

The tractability landscape of diffusion alignment: regularization, rewards, and computational primitives

Inference-time reward alignment asks how to turn a pre-trained diffusion model with base law

p

into a sampler that favors a reward

r

while remaining close to

p

. Since there is no canonical distributional distance for this closeness constraint, different choices lead to different "reward-aligned" laws and, just as importantly, different algorithmic problems. We develop a primitive-based approach to reward alignment: rather than assuming arbitrary reward-aligned laws can be sampled, we ask which simple algorithmic primitives suffice to implement alignment for non-trivial reward classes. If closeness is measured in KL distance, the target law is

q(x) \propto p(x) \exp(λ^{-1}r(x))

. For this setting, we show that linear exponential tilts of the form

q(x)\propto p(x)\exp(\langle θ, x \rangle)

-- which according to recent work [MRR26] can be efficiently sampled from -- are a sufficient primitive for aligning to a very broad class of convex low-dimensional rewards. If closeness is measured in Wasserstein distance, the corresponding primitive is a proximal transport oracle: given

x

, solve

\mbox{argmax}_y \{r(y)- λc(x,y)\}

. This oracle can be efficiently implemented for concave or low-dimensional Lipschitz rewards

r(x)=f(Ax)

. Together, these results illustrate that the choice of distribution distance for alignment affects the computational primitive and the tractable reward class.

Ankur Moitra, Andrej Risteski, Dhruv Rohatgi

Diffusion Alignment Reward Gradients Regularization Tractability

May 12, 2026cs.LG

Gradient-Free Noise Optimization for Reward Alignment in Generative Models

Existing reward alignment methods for diffusion and flow models rely on multi-step stochastic trajectories, making them difficult to extend to deterministic generators. A natural alternative is noise-space optimization, but existing approaches require backpropagation through the generator and reward pipeline, limiting applicability to differentiable settings. To address this, here we present ZeNO (Zeroth-order Noise Optimization), a gradient-free framework that formulates noise optimization as a path-integral control problem, estimable from zeroth-order reward evaluations alone. When instantiated with an Ornstein--Uhlenbeck reference process, the update connects to Langevin dynamics implicitly targeting a reward-tilted distribution. ZeNO enables effective inference-time scaling and demonstrates strong performance across diverse generators and reward functions, including a protein structure generation task where backpropagation is infeasible.

Jeongsol Kim, Hongeun Kim, Jian Wang +1

KAIST AI1 · Snap Inc.2

Reward Gradients Diffusion Alignment Gradient-Based Optimization Generative Models Langevin Dynamics

May 10, 2026cs.CV

Offline Preference Optimization for Rectified Flow with Noise-Tracked Pairs

Existing preference datasets for text-to-image models typically store only the final winner/loser images. This representation is insufficient for rectified flow (RF) models, whose generation is naturally indexed by a specific prior noise sample and follows a nearly straight denoising trajectory. In contrast, prior DPO-style alignment for diffusion models commonly estimates trajectories using an independent forward noising process, which can be mismatched to the true reverse dynamics and introduces unnecessary variance. We propose Prior Noise-Aware Preference Optimization (PNAPO), an off-policy alignment framework specialized for rectified flow. PNAPO augments preference data by retaining the paired prior noises used to generate each winner/loser image, turning the standard (prompt, winner, loser) triplet into a sextuple. Leveraging the straight-line property of RF, we estimate intermediate states via noise-image interpolation, which constrains the trajectory estimation space and yields a tighter surrogate objective for preference optimization. In addition, we introduce a dynamic regularization strategy that adapts the DPO regularization based on (i) the reward gap between winner and loser and (ii) training progress, improving stability and sample efficiency. Experiments on state-of-the-art RF T2I backbones show that PNAPO consistently improves preference metrics while substantially reducing training compute.

Yunhong Lu, Qichao Wang, Hengyuan Cao +2

Zhejiang University · Shanghai Institute for Advanced Study-Zhejiang University · Shanghai Institute for Mathematics and Interdisciplinary Sciences

Rectified Flow Diffusion Alignment Text-To-Image Diffusion Models Denoising Trajectory Text-To-Image

May 10, 2026cs.LG

DiffATS: Diffusion in Aligned Tensor Space

Direct diffusion modeling of high-resolution spatiotemporal fields is computationally challenging. Parameter-efficient primitives address this by representing high-dimensional data with a compact set of parameters. In this paper, we construct data-dependent tensor primitives without pretrained compression autoencoders. Our construction starts from Tucker decomposition, which captures low-rank multilinear structure through a core tensor and mode-wise factors. However, Tucker factors are non-unique: the same tensor can be represented by different rotated factors, which complicates generative modeling. We address this issue with orthogonal Procrustes (OP) alignment. Specifically, we select medoid anchor matrices from the data and align the factor matrices to resolve the gauge ambiguity. This yields matrix Grassmannian primitives and tensor Grassmannian primitives that are compact, data-adaptive, and directly decodable by explicit multilinear reconstruction. Theoretically, we prove that the proposed primitive maps are homeomorphisms between low-rank tensors and their corresponding primitive spaces, certifying that the representations are non-degenerate and topologically faithful. Building on these primitives, we propose Diffusion in Aligned Tensor Space (DiffATS), a generative framework that trains diffusion models directly on aligned tensor primitives. Across images, videos, and PDE solutions, DiffATS achieves strong unconditional and conditional generation performance while compressing original data by

3.9\times

to

210\times

, without relying on any pretrained deep compression autoencoders.

Jinhua Lyu, Tianmin Yu, Brian Kim +3

Northwestern University

Diffusion Alignment Diffusion Models Autoencoder Architectures Spatiotemporal Fields

May 9, 2026cs.LG

TMPO: Trajectory Matching Policy Optimization for Diverse and Efficient Diffusion Alignment

Reinforcement learning (RL) has shown extraordinary potential in aligning diffusion models to downstream tasks, yet most of them still suffer from significant reward hacking, which degrades generative diversity and quality by inducing visual mode collapse and amplifying unreliable rewards. We identify the root cause as the mode-seeking nature of these methods, which maximize expected reward without effectively constraining probability distribution over acceptable trajectories, causing concentration on a few high-reward paths. In contrast, we propose Trajectory Matching Policy Optimization (TMPO), which replaces scalar reward maximization with trajectory-level reward distribution matching. Specifically, TMPO introduces a Softmax Trajectory Balance (Softmax-TB) objective to match the policy probabilities of K trajectories to a reward-induced Boltzmann distribution. We prove that this objective inherits the mode-covering property of forward KL divergence, preserving coverage over all acceptable trajectories while optimizing reward. To further reduce multi-trajectory training time on large-scale flow-matching models, TMPO incorporates Dynamic Stochastic Tree Sampling, where trajectories share denoising prefixes and branch at dynamically scheduled steps, reducing redundant computation while improving training effectiveness. Extensive results across diverse alignment tasks such as human preference, compositional generation and text rendering show that TMPO improves generative diversity over state-of-the-art methods by 9.1%, and achieves competitive performance in all downstream and efficiency metrics, attaining the optimal trade-off between reward and diversity.

Jiaming Li, Chenyu Zhu, Nanxi Yi +9

1MAIR Lab, Huazhong University of Science and Technology · 2Kuaishou Technology · 3Nanyang Technological University +2

Diffusion Alignment Diffusion-Based Reinforcement Learning Methods Reward Functions Diversity Boltzmann Distributions

May 8, 2026cs.CV

Flow-OPD: On-Policy Distillation for Flow Matching Models

Existing Flow Matching (FM) text-to-image models suffer from two critical bottlenecks under multi-task alignment: the reward sparsity induced by scalar-valued rewards, and the gradient interference arising from jointly optimizing heterogeneous objectives, which together give rise to a 'seesaw effect' of competing metrics and pervasive reward hacking. Inspired by the success of On-Policy Distillation (OPD) in the large language model community, we propose Flow-OPD, the first unified post-training framework that integrates on-policy distillation into Flow Matching models. Flow-OPD adopts a two-stage alignment strategy: it first cultivates domain-specialized teacher models via single-reward GRPO fine-tuning, allowing each expert to reach its performance ceiling in isolation; it then establishes a robust initial policy through a Flow-based Cold-Start scheme and seamlessly consolidates heterogeneous expertise into a single student via a three-step orchestration of on-policy sampling, task-routing labeling, and dense trajectory-level supervision. We further introduce Manifold Anchor Regularization (MAR), which leverages a task-agnostic teacher to provide full-data supervision that anchors generation to a high-quality manifold, effectively mitigating the aesthetic degradation commonly observed in purely RL-driven alignment. Built upon Stable Diffusion 3.5 Medium, Flow-OPD raises the GenEval score from 63 to 92 and the OCR accuracy from 59 to 94, yielding an overall improvement of roughly 10 points over vanilla GRPO, while preserving image fidelity and human-preference alignment and exhibiting an emergent 'teacher-surpassing' effect. These results establish Flow-OPD as a scalable alignment paradigm for building generalist text-to-image models. The codes and weights will be released in: https://github.com/CostaliyA/Flow-OPD .

Zhen Fang, Wenxuan Huang, Yu Zeng +8

University of Science and Technology of China · University of California, Los Angeles · 3The Chinese University of Hong Kong +1

Text-To-Image Diffusion Models Image-Text Alignment Diffusion Alignment Text-To-Image Uni-Opd Post-Training

May 8, 2026cs.CV

Diffusion-APO: Trajectory-Aware Direct Preference Alignment for Video Diffusion Transformers

Efficiently aligning large-scale video diffusion models with human intent requires a scalable and trajectory-aware pathway that bridges the inherent discrepancy between training noise distributions and practical inference trajectories. While existing paradigms such as Direct Preference Optimization (DPO) and Group Relative Policy Optimization (GRPO) attempt to address this, they are often hindered by either reliance on bias-prone, complex reward models or suboptimal timestep sampling. In this paper, we propose Diffusion-APO (Aligned Preference Optimization), a trajectory-aware algorithm that resolves this misalignment by synchronizing training noise with inference-time denoising paths to maximize gradient signal efficacy. To translate this algorithmic innovation into a practical solution, we introduce a unified and modular RLHF framework that integrates online ranking, half-online anchoring, offline refinement, and distillation-aware drift correction. This framework enables flexible, multi-stage preference alignment across diverse data and computational constraints without relying on scalar-reward-based policy gradients. Through extensive experiments, we demonstrate that Diffusion-APO consistently outperforms standard baselines in visual quality and instruction following, while effectively preserving generative fidelity during model acceleration, providing a robust, end-to-end pathway for scalable video diffusion alignment.

Jingyuan Zhu, Biaolong Chen, Le Zhang +3

Alibaba Group

Diffusion Alignment Video Diffusion Models Diffusion-Based Reinforcement Learning Methods Reward-Only Policy Optimization Variants

May 8, 2026cs.LG

Inference-Time Attribute Distribution Alignment for Unconditional Diffusion

Inference-time controllable generation is essential for real-world applications of unconditional diffusion models. However, most existing techniques focus on individual samples, struggling in applications that require the sample population to follow specific attribute distributions (e.g., demographic balance or semantic proportions). We formalize this setting as the inference-time attribute distributional alignment problem for pretrained unconditional diffusion models. To address this, we cast inference-time attribute distributional alignment as an optimal control problem over the reverse diffusion process, viewing the process as the rollout of a dynamical system and augmenting it with additive, time-dependent perturbations as control. We solve for the perturbations using an optimal-control-based algorithm to optimize a differentiable distribution-matching objective while penalizing control effort to preserve data fidelity. Experiment results in image generation demonstrate that our proposed plug-and-play approach can better align attribute distributions to diverse and flexible test-time targets compared to baselines, without retraining or finetuning the pretrained diffusion model.

Hao Luan, See-Kiong Ng, Chun Kai Ling

School of Computing, National University of Singapore · Institute of Data Science, National University of Singapore

Diffusion Alignment Diffusion Models Generative Models Distributional Alignment Distributions

May 8, 2026cs.LG

UMEDA: Unified Multi-modal Efficient Data Fusion for Privacy-Preserving Graph Federated Learning via Spectral-Gated Attention and Diffusion-Based Operator Alignment

Device-free localization trains models from heterogeneous wireless and visual sensors (e.g., Wi-Fi, LiDAR) distributed across edge devices. Federated learning offers a privacy-respecting framework, but is brittle when clients differ in sensor modality and resolution, when their data distributions drift, and when privacy noise destroys the structural signal needed for localization. We propose UMEDA, a graph federated learning framework in which clients form nodes of a global graph that share a continuous integral operator, and aggregation is reformulated as spectral signal processing on this operator. Each client encodes its local sensors with a linear-attention layer whose kernel spectrum is low-rank filtered, suppressing modality-specific residuals so clients with different sensors align in a common low-rank subspace. The server then aggregates client updates via a diffusion model over the kernel's spectral coefficients, treating updates as discretizations of a shared operator rather than topology-bound weights -- this absorbs varying graph sizes and missing modalities without node-wise correspondence. To balance privacy and utility, we add an anisotropic differential-privacy mechanism that projects noise preferentially into the null space of the signal subspace, preserving dominant eigendirections while ensuring formal

(ε, δ)

-DP under gradient clipping. On MM-Fi and the RELI11D out-of-distribution benchmark, UMEDA outperforms state-of-the-art federated baselines in accuracy, convergence, and communication efficiency, particularly under high modality heterogeneity and tight privacy budgets.

Shih-Yu Lai, Hirozumi Yamaguchi, Shang-Tse Chen +2

National Taiwan University · RIKEN-CCS · The University of Osaka +1

Federated Learning Multimodal Fusion Diffusion Alignment Cross-View Geo-Localization

May 8, 2026stat.ML

TRACE: Transport Alignment Conformal Prediction via Diffusion and Flow Matching Models

Constructing valid and informative conformal prediction regions for multi-dimensional outputs remains a fundamental challenge. While conformal prediction provides finite-sample, distribution-free coverage guarantees, its practical performance critically depends on the choice of nonconformity score. Existing approaches often rely on restrictive geometric assumptions or require explicit likelihood evaluation and invertible transformations, limiting their applicability in complex generative settings. In this work, we introduce TRACE (TRansport Alignment Conformal Estimation), a conformal prediction framework that defines nonconformity through transport alignment in diffusion and flow matching models. Rather than evaluating likelihoods, we measure how well a candidate output aligns with the learned generative dynamics by averaging denoising or velocity-matching errors along stochastic transport trajectories. The resulting transport-based scores are scalar-valued and can be calibrated using split conformal prediction, yielding valid marginal coverage under exchangeability. We further analyze the statistical properties of the proposed scores and their sensitivity to computational budget. Experiments on synthetic and real datasets demonstrate valid coverage and show that the resulting regions adapt naturally to multimodal and non-convex conditional distributions.

Zhenhan Fang, Aixin Tan, Jian Huang

Department of Statistics and Actuarial Science University of Iowa Iowa, IA 52242, USA · Departments of Data Science and AI, and Applied Mathematics The Hong Kong Polytechnic University Hong Kong, China

Online Conformal Prediction Nonconformity Scores Diffusion Alignment Distributional Fidelity Transport

May 7, 2026cs.CV

MARBLE: Multi-Aspect Reward Balance for Diffusion RL

Reinforcement learning fine-tuning has become the dominant approach for aligning diffusion models with human preferences. However, assessing images is intrinsically a multi-dimensional task, and multiple evaluation criteria need to be optimized simultaneously. Existing practice deal with multiple rewards by training one specialist model per reward, optimizing a weighted-sum reward

R(x)=\sum_k w_k R_k(x)

, or sequentially fine-tuning with a hand-crafted stage schedule. These approaches either fail to produce a unified model that can be jointly trained on all rewards or necessitates heavy manually tuned sequential training. We find that the failure stems from using a naive weighted-sum reward aggregation. This approach suffers from a sample-level mismatch because most rollouts are specialist samples, highly informative for certain reward dimensions but irrelevant for others; consequently, weighted summation dilutes their supervision. To address this issue, we propose MARBLE (Multi-Aspect Reward BaLancE), a gradient-space optimization framework that maintains independent advantage estimators for each reward, computes per-reward policy gradients, and harmonizes them into a single update direction without manually-tuned reward weighting, by solving a Quadratic Programming problem. We further propose an amortized formulation that exploits the affine structure of the loss used in DiffusionNFT, to reduce the per-step cost from K+1 backward passes to near single-reward baseline cost, together with EMA smoothing on the balancing coefficients to stabilize updates against transient single-batch fluctuations. On SD3.5 Medium with five rewards, MARBLE improves all five reward dimensions simultaneously, turns the worst-aligned reward's gradient cosine from negative under weighted summation in 80% of mini-batches to consistently positive, and runs at 0.97X the training speed of baseline training.

Canyu Zhao, Hao Chen, Yunze Tong +3

Zhejiang University · Work done during an internship at HiThink. · HiThink +1

Diffusion-Based Reinforcement Learning Methods Diffusion Alignment Reward Gradients Diffusion Models Cosine Similarity

May 6, 2026cs.LG

Towards General Preference Alignment: Diffusion Models at Nash Equilibrium

Reinforcement learning from human feedback (RLHF) has been popular for aligning text-to-image (T2I) diffusion models with human preferences. As a mainstream branch of RLHF, Direct Preference Optimization (DPO) offers a computationally efficient alternative that avoids explicit reward modeling and has been widely adopted in diffusion alignment. However, existing preference-based methods for diffusion alignment still rely on reward-induced preference signals and typically assume that human preferences can be adequately modeled by the Bradley--Terry (BT) model, which may fail to capture the full complexity of human preferences. In this paper, we formulate diffusion alignment from a game-theoretic perspective. We propose Diffusion Nash Preference Optimization (Diff.-NPO), an intuitive general preference framework for diffusion alignment. Diff.-NPO encourages the current policy to play against itself to achieve self improvement and lead to a better alignment. Empirically, we demonstrate the effectiveness of Diff.-NPO on the text-to-image generation task via various metrics. Diff.-NPO consistently outperforms existing preference-based diffusion alignment methods.

Jiaming Hu, Jiamu Bai, Haoyu Wang +2

Boston University · Penn State University · University at Albany

Diffusion Alignment Preference Alignment Diffusion Models

May 5, 2026cs.CV

AHPA: Adaptive Hierarchical Prior Alignment for Diffusion Transformers

Representation alignment has recently emerged as an effective paradigm for accelerating Diffusion Transformer training. Despite their success, existing alignment methods typically impose a fixed supervision target or a fixed alignment granularity throughout the entire denoising trajectory, whether the guidance is provided by external vision encoders, internal self-representations, or VAE-derived features. We argue that such timestep-agnostic alignment is suboptimal because the useful granularity of representation supervision changes systematically with the signal-to-noise ratio. In high-noise regimes, diffusion models benefit more from coarse semantic and layout-level anchoring, whereas in low-noise regimes, the training signal should emphasize spatially detailed and structurally faithful refinement. This non-stationary alignment behavior creates a representational mismatch for static single-level supervisors. To address this issue, we propose Adaptive Hierarchical Prior Alignment (AHPA), a lightweight alignment framework that exploits the hierarchical representations naturally embedded in the frozen VAE encoder. Instead of using only a single compressed latent as the alignment target, AHPA extracts multi-level VAE features that provide complementary priors ranging from local geometry and spatial topology to coarse semantic layout. A timestep-conditioned Dynamic Router adaptively selects and weights these hierarchical priors along the denoising trajectory, thereby synchronizing the alignment granularity with the model's evolving training needs. Extensive experiments show that AHPA improves convergence and generation quality over baselines and incurs no additional inference cost while avoiding external encoder supervision during training.

Ruibin Min, Yexin Liu, Aimin Pan +5

Sun Yat-sen University · The Hong Kong University of Science and Technology · VNET Group +2

Diffusion Alignment Diffusion Transformers Denoising Trajectory Conditional Variational Autoencoder Representational Collapse

Apr 27, 2026cs.CV

Learning from Noisy Preferences: A Semi-Supervised Learning Approach to Direct Preference Optimization

Human visual preferences are inherently multi-dimensional, encompassing aesthetics, detail fidelity, and semantic alignment. However, existing datasets provide only single, holistic annotations, resulting in severe label noise: images that excel in some dimensions but are deficient in others are simply marked as winner or loser. We theoretically demonstrate that compressing multi-dimensional preferences into binary labels generates conflicting gradient signals that misguide Diffusion Direct Preference Optimization (DPO). To address this, we propose Semi-DPO, a semi-supervised approach that treats consistent pairs as clean labeled data and conflicting ones as noisy unlabeled data. Our method starts by training on a consensus-filtered clean subset, then uses this model as an implicit classifier to generate pseudo-labels for the noisy set for iterative refinement. Experimental results demonstrate that Semi-DPO achieves state-of-the-art performance and significantly improves alignment with complex human preferences, without requiring additional human annotation or explicit reward models during training. We will release our code and models at: https://github.com/L-CodingSpace/semi-dpo

Xinxin Liu, Ming Li, Zonglin Lyu +2

University of Central Florida

Backtranslation Augmented Direct Preference Optimization Diffusion Alignment Semi-Supervised Learning Human Annotations Visual Perception

Apr 26, 2026cs.CV

Oracle Noise: Faster Semantic Spherical Alignment for Interpretable Latent Optimization

Text-to-image diffusion models have achieved remarkable generative capabilities, yet accurately aligning complex textual prompts with synthesized layouts remains an ongoing challenge. In these models, the initial Gaussian noise acts as a critical structural seed dictating the macroscopic layout. Recent online optimization and search methods attempt to refine this noise to enhance text-image alignment. However, relying on unconstrained Euclidean gradient ascent mathematically inflates the latent norm and destroys the standard Gaussian prior, causing severe visual artifacts like color over-saturation. Furthermore, these methods suffer from inefficient semantic routing and easily fall into the ``reward hacking'' trap of external proxy models. To address these intertwined bottlenecks, we propose Oracle Noise, a zero-shot framework reframing noise initialization as semantic-driven optimization strictly confined to a Riemannian hypersphere. Instead of relying on complex external parsers, we directly identify the most impactful structural words in the prompt to efficiently route optimization energy. By updating the noise strictly along a spherical path, we mathematically preserve the original Gaussian distribution. This geometric constraint eliminates norm inflation and unlocks aggressive step sizes for rapid convergence. Extensive experiments demonstrate that Oracle Noise significantly accelerates semantic alignment and achieves superior aesthetics without black-box models. It completely mitigates Euclidean-induced degradation, establishing state-of-the-art performance across human preference metrics (e.g., HPSv2, ImageReward), semantic alignment (CLIP Score), and sample diversity, all within a strict 2-second optimization budget.

Haosen Li, Wenshuo Chen, Lei Wang +3

1The Hong Kong University of Science and Technology (Guangzhou) · 2Griffith University & Data61/CSIRO

Diffusion Alignment Text-To-Image Diffusion Models Image-Text Alignment Layouts Oracles Latent Variable

Apr 23, 2026eess.IV

Conditional Diffusion Posterior Alignment for Sparse-View CT Reconstruction

Computed Tomography (CT) is a widely used imaging modality in medical and industrial applications. To limit radiation exposure and measurement time, there is a growing interest in sparse-view CT, where the number of projection views is significantly reduced. Deep neural networks have shown great promise in improving reconstruction quality in sparse-view CT, especially generative diffusion models. However, these methods struggle to scale to large 3D volumes due to several reasons: (i) the high memory and computational requirements of 3D models, (ii) the lack of large 3D training datasets, and (iii) the inconsistencies across slices when using 2D models independently on each slice. We overcome these limitations and scale diffusion-based sparse-view CT reconstruction to large 3D volumes by combining conditional diffusion with explicit data consistency. We propose Conditional Diffusion Posterior Alignment (CDPA) to enable scalable 3D sparse-view CT reconstruction. A 2D U-Net diffusion model is conditioned on an initial 3D reconstruction to improve inter-slice consistency, combined with data-consistency alignment to match measured projections. Experiments on synthetic and real Cone Beam CT (CBCT) data show state-of-the-art performance, with ablations that confirm the synergistic effects of the proposed pipeline. Finally, we show that the same principles also strengthen fast denoising U-Nets, yielding near-diffusion quality at a fraction of the computational cost.

Luis Barba, Johannes Kirschner, Benjamin Bejar

Swiss Data Science Center (SDSC) in Paul Scherrer Institute (PSI), Villigen, Switzerland. · Swiss Data Science Center (SDSC) and ETH Zurich, Switzerland. · Swiss Data Science Center (SDSC) and Paul Scherrer Institute (PSI), Villigen, Switzerland.

Cone-Beam Computed Tomography Sparse-View Diffusion Priors U-Net Diffusion Alignment Diffusion Models

Apr 19, 2026cs.LG

Reward Score Matching: Unifying Reward-based Fine-tuning for Flow and Diffusion Models

Reward-based fine-tuning steers a pretrained diffusion or flow-based generative model toward higher-reward samples while remaining close to the pretrained model. Although existing methods are derived from different perspectives, we show that many can be written under a common framework, which we call reward score matching (RSM). Under this view, alignment becomes score matching against a value-guided target, and the main differences across methods reduce to the construction of the value-guidance estimator and the effective optimization strength across timesteps. This unification clarifies the bias-variance-compute tradeoffs of existing designs, and distinguishes core optimization components from auxiliary mechanisms that add complexity without clear benefit. Guided by this perspective, we develop simpler, more efficient redesigns across representative differentiable and black-box reward alignment tasks. Overall, RSM turns a seemingly fragmented collection of reward-based fine-tuning methods into a smaller, more interpretable, and more actionable design space. Code is available at https://github.com/jaylee2000/rsm

Jeongjae Lee, Jinho Chang, Jeongsol Kim +1

Graduate School of AI KAIST, Korea

Diffusion Alignment Diffusion Models Generative Flow Networks Reinforcement Fine-Tuning

Mar 18, 2026cs.CV

SHIFT: Motion Alignment in Video Diffusion Models with Adversarial Hybrid Fine-Tuning

Image-conditioned video diffusion models achieve impressive visual realism but often suffer from weakened motion fidelity, e.g., reduced motion dynamics or degraded long-term temporal coherence, especially after fine-tuning. We study motion alignment in video diffusion models post-training. To address this, we introduce pixel-motion rewards based on pixel flux dynamics, capturing both instantaneous and long-term motion consistency. We further propose \underline{S}mooth \underline{H}ybr\underline{i}d \underline{F}ine-\underline{t}uning (SHIFT), a scalable reward-driven framework that unifies supervised fine-tuning and advantage-weighted fine-tuning. Benefiting from novel adversarial advantages, SHIFT improves convergence speed and mitigates reward hacking. Experiments show that our approach efficiently resolves dynamic-degree collapse in modern video diffusion models supervised fine-tuning. Project page: https://xiye20.github.io/projects/SHIFT/.

Xi Ye, Wenjia Yang, Yangyang Xu +4

Tsinghua University · Ant Group · University of Chinese Academy of Sciences

Video Diffusion Models Diffusion Alignment Adversarial Training Long-Term