Denoising
-
RefDecoder: Enhancing Visual Generation with Conditional Video Decoding
2026-05-14 17:59:52Xiang Fan, Yuheng Wang, Bohan Fang, Zhongzheng Ren, Ranjay KrishnaAbstract
Video generation powers a vast array of downstream applications. However, while the de facto standard, i.e., latent diffusion models, typically employ heavily conditioned denoising networks, their decoders often remain unconditional. We observe that this architectural asymmetry leads to significant loss of detail and inconsistency relative to the input image. To address this, we argue that the decoder requires equal conditioning to preserve structural integrity. We introduce RefDecoder, a reference-conditioned video VAE decoder by injecting high-fidelity reference image signal directly into the decoding process via reference attention. Specifically, a lightweight image encoder maps the reference frame into the detail-rich high-dimensional tokens, which are co-processed with the denoised video latent tokens at each decoder up-sampling stage. We demonstrate consistent improvements across several distinct decoder backbones (e.g., Wan 2.1 and VideoVAE+), achieving up to +2.1dB PSNR over the unconditional baselines on the Inter4K, WebVid, and Large Motion reconstruction benchmarks. Notably, RefDecoder can be directly swapped into existing video generation systems without additional fine-tuning, and we report across-the-board improvements in subject consistency, background consistency, and overall quality scores on the VBench I2V benchmark. Beyond I2V, RefDecoder generalizes well to a wide range of visual generation tasks such as style transfer and video editing refinement.
Abstract (translated)
URL
https://arxiv.org/abs/2605.15196
PDF
-
RAVEN: Real-time Autoregressive Video Extrapolation with Consistency-model GRPO
2026-05-14 17:59:30Yanzuo Lu, Ronglai Zuo, Jiankang DengAbstract
Causal autoregressive video diffusion models support real-time streaming generation by extrapolating future chunks from previously generated content. Distilling such generators from high-fidelity bidirectional teachers yields competitive few-step models, yet a persistent gap between the history distributions encountered during training and those arising at inference constrains generation quality over long horizons. We introduce the Real-time Autoregressive Video Extrapolation Network (RAVEN), a training-time test framework that repacks each self rollout into an interleaved sequence of clean historical endpoints and noisy denoising states. This formulation aligns training attention with inference-time extrapolation and allows downstream chunk losses to supervise the history representations on which future predictions depend. We further propose Consistency-model Group Relative Policy Optimization (CM-GRPO), which reformulates a consistency sampling step as a conditional Gaussian transition and applies online Reinforcement Learning (RL) directly to this kernel, avoiding the Euler-Maruyama auxiliary process adopted in prior flow-model RL formulations. Experiments demonstrate that RAVEN surpasses recent causal video distillation baselines across quality, semantic, and dynamic degree evaluations, and that CM-GRPO provides further gains when combined with RAVEN.
Abstract (translated)
URL
https://arxiv.org/abs/2605.15190
PDF
-
Warp-as-History: Generalizable Camera-Controlled Video Generation from One Training Video
2026-05-14 17:58:26Yifan Wang, Tong HeAbstract
Camera-controlled video generation has made substantial progress, enabling generated videos to follow prescribed viewpoint trajectories. However, existing methods usually learn camera-specific conditioning through camera encoders, control branches, or attention and positional-encoding modifications, which often require post-training on large-scale camera-annotated videos. Training-free alternatives avoid such post-training, but often shift the cost to test-time optimization or extra denoising-time guidance. We propose Warp-as-History, a simple interface that turns camera-induced warps into camera-warped pseudo-history with target-frame positional alignment and visible-token selection. Given a target camera trajectory, we construct camera-warped pseudo-history from past observations and feed it through the model's visual-history pathway. Crucially, we align its positional encoding with the target frames being denoised and remove warped-history tokens without valid source observations. Without any training, architectural modification, or test-time optimization, this interface reveals a non-trivial zero-shot capability of a frozen video generation model to follow camera trajectories. Moreover, lightweight offline LoRA finetuning on only one camera-annotated video further improves this capability and generalizes to unseen videos, improving camera adherence, visual quality, and motion dynamics without test-time optimization or target-video adaptation. Extensive experiments on diverse datasets confirm the effectiveness of our method.
Abstract (translated)
URL
https://arxiv.org/abs/2605.15182
PDF
-
Pelican-Unified 1.0: A Unified Embodied Intelligence Model for Understanding, Reasoning, Imagination and Action
2026-05-14 17:50:42Yi Zhang, Yinda Chen, Che Liu, Zeyuan Ding, Jin Xu, Shilong Zou, Junwei Liao, Jiayu Hu, Xiancong Ren, Xiaopeng Zhang, Yechi Liu, Haoyuan Shi, Zecong Tang, Haosong Sun, Renwen Cui, Kuishu Wu, Wenhai Liu, Yang Xu, Yingji Zhang, Yidong Wang, Senkang Hu, Jinpeng Lu, Nga Teng Chan, Yechen Wu, Yong Dai, Jian Tang, Xiaozhu JuAbstract
We present Pelican-Unified 1.0, the first embodied foundation model trained according to the principle of unification. Pelican-Unified 1.0 uses a single VLM as a unified understanding module, mapping scenes, instructions, visual contexts, and action histories into a shared semantic space. The same VLM also serves as a unified reasoning module, autoregressively producing task-, action-, and future-oriented chains of thought in a single forward pass and projecting the final hidden state into a dense latent variable. A Unified Future Generator (UFG) then conditions on this latent variable and jointly generates future videos and future actions through two modality-specific output heads within the same denoising process. The language, video, and action losses are all backpropagated into the shared representation, enabling the model to jointly optimize understanding, reasoning, imagination, and action during training, rather than training three isolated expert systems. Experiments demonstrate that unification does not imply compromise. With a single checkpoint, Pelican-Unified 1.0 achieves strong performance across all three capabilities: 64.7 on eight VLM benchmarks, the best among comparable-scale models; 66.03 on WorldArena, ranking first; and 93.5 on RoboTwin, the second-best average among compared action methods. These results show that the unified paradigm succeeds in preserving specialist strength while bringing understanding, reasoning, imagination, and action into one model.
Abstract (translated)
URL
https://arxiv.org/abs/2605.15153
PDF
-
Compositional Video Generation via Inference-Time Guidance
2026-05-14 15:50:25Ariel Shaulov, Eitan Shaar, Amit Edenzon, Gal Chechik, Lior WolfAbstract
Text-to-video diffusion models generate realistic videos, but often fail on prompts requiring fine-grained compositional understanding, such as relations between entities, attributes, actions, and motion directions. We hypothesize that these failures need not be addressed by retraining the generator, but can instead be mitigated by steering the denoising process using the model's own internal grounding signals. We propose \textbf{CVG}, an inference-time guidance method for improving compositional faithfulness in frozen text-to-video models. Our key observation is that cross-attention maps already encode how prompt concepts are grounded across space and time. We train a lightweight compositional classifier on these attention features and use its gradients during early denoising steps to steer the latent trajectory toward the desired composition. Built on a frozen VLM backbone, the classifier transfers across semantically related composition labels rather than relying only on narrow category-specific features. CVG improves compositional generation without modifying the model architecture, fine-tuning the generator, or requiring layouts, boxes, or other user-supplied controls. Experiments on compositional text-to-video benchmarks show improved prompt faithfulness while preserving the visual quality of the underlying generator.
Abstract (translated)
URL
https://arxiv.org/abs/2605.14988
PDF
-
Multi-scale Coarse-to-fine Modeling for Test-time Human Motion Control
2026-05-14 15:09:49Nhat Le, Daochang Liu, Anh Nguyen, Ajmal MianAbstract
We present MSCoT, a multi-scale, coarse-to-fine model for test-time human motion synthesis and control. Unlike recent approaches that rely on multiple iterative denoising/token-prediction steps, or modules tailored for specific control signals, MSCoT discretizes motion into a multi-scale hierarchical representation and predicts the entire token sequence at each temporal scale in a coarse-to-fine fashion. Building on this coarse-to-fine paradigm, we propose an efficient multi-scale token guidance strategy that overcomes the challenge of discrete sampling and steers the token distribution towards the control goals, allowing for fast and flexible control. To address the limitations of a discrete codebook, a lightweight token refiner further adds continuous residuals to the discrete token embeddings and allows differentiable test-time refinement optimization to ensure precise alignment with the control objectives. MSCoT is able to produce quality motions, consistent with the control constraints, while offering substantially faster sampling than diffusion-based approaches. Experiments on popular benchmarks demonstrate state-of-the-art controllable text-to-motion generation performance of MSCoT over existing baselines, with better motion quality (48% FID improvement), higher control accuracy (-61% avg error), and $10 \times$ faster inference speed on HumanML3D.
Abstract (translated)
URL
https://arxiv.org/abs/2605.14935
PDF
-
SEDiT: Mask-Free Video Subtitle Erasure via One-step Diffusion Transformer
2026-05-14 14:37:54Zheng Hui, Yunlong BaiAbstract
Recent breakthroughs in video diffusion models have significantly accelerated the development of video editing techniques. However, existing methods often rely on inpainting video frames based on masked input, which requires extracting the target video mask in advance, and the precision of the segmentation directly affects the quality of the completion. In this paper, we present SEDiT, a novel one-stage video Subtitle Erasure approach via One-step Diffusion Transformer. We introduce a mask-free inference approach that enables direct erasure of the targeted subtitle. The proposed one-stage framework mitigates the sub-optimality inherent in the two-stage processing of prior models. Since subtitle removal is a localized editing task in which most pixels remain unchanged, the underlying distribution shift is minimal, making it well-suited to one-step generation under rectified flow. We empirically validate the reliability of one-step denoising and further provide a formal theoretical justification. Under the localized-editing structure of subtitle removal, the conditional optimal transport (OT) map and its induced rectified flow velocity field are Lipschitz continuous with respect to the latent variable, which underpins the theoretical feasibility of one-step sampling. To address the challenge of long-term temporal consistency, we adopt a hybrid training strategy by occasionally conditioning the model with a clean first-frame latent. This facilitates temporal continuity, allowing each segment during inference to leverage the output of its predecessor. To avoid visible seams caused by cropping and reinserting processed targets, particularly in scenarios involving substantial motion, we feed the original video directly into SEDiT. Thanks to one-step and chunk-wise streaming inference, our method can efficiently handle native 1440p video with infinite length.
Abstract (translated)
URL
https://arxiv.org/abs/2605.14894
PDF
-
Denoising-GS: Gaussian Splatting with Spatial-aware Denoising
2026-05-14 14:24:21Qingyuan Zhou, Xinyi Liu, Weidong Yang, Ning Wang, Shuquan Ye, Ben Fei, Ying He, Wanli OuyangAbstract
Recent advances in 3D Gaussian Splatting (3DGS) have achieved remarkable success in high-fidelity Novel View Synthesis (NVS), yet the optimization process inevitably introduces noisy Gaussian primitives due to the sparse and incomplete initialization from Structure-from-Motion (SfM) point clouds. Most existing methods focus solely on adjusting the positions of primitives during optimization, while neglecting the underlying spatial structure. To this end, we introduce a new perspective by formulating the optimization of 3DGS as a primitive denoising process and propose Denoising-GS, a spatial-aware denoising framework for Gaussian primitives by taking both the positions and spatial structure into consideration. Specifically, we design an optimizer that preserves the spatial optimization flow of primitives, facilitating coherent and directed denoising rather than random perturbations. Building upon this, the Spatial Gradient-based Denoising strategy jointly considers the spatial supports of primitives to ensure gradient-consistent updates. Furthermore, the Uncertainty-based Denoising module estimates primitive-wise uncertainty to prune redundant or noisy primitives, while the Spatial Coherence Refinement strategy selectively splits primitives in sparse regions to maintain structural completeness. Experiments conducted on three benchmark datasets demonstrate that Denoising-GS consistently enhances NVS fidelity while maintaining representation compactness, achieving state-of-the-art performance across all benchmarks. Source code and models will be made publicly available.
Abstract (translated)
URL
https://arxiv.org/abs/2605.14880
PDF
-
Unlocking Complex Visual Generation via Closed-Loop Verified Reasoning
2026-05-14 14:22:25Hanbo Cheng, Limin Lin, Ruo Zhang, Yicheng Pan, Jun DuAbstract
Despite rapid advancements, current text-to-image (T2I) models predominantly rely on a single-step generation paradigm, which struggles with complex semantics and faces diminishing returns from parameter scaling. While recent multi-step reasoning approaches show promise, they are hindered by ungrounded planning hallucinations lacking verification, monolithic post-hoc reflection, long-context optimization instabilities, and prohibitive inference latency. To overcome these bottlenecks, we propose the Closed-Loop Visual Reasoning (CLVR) framework, a comprehensive system that deeply couples visual-language logical planning with pixel-level diffusion generation. CLVR introduces an automated data engine with step-level visual verification to synthesize reliable reasoning trajectories, and proposes Proxy Prompt Reinforcement Learning (PPRL) to resolve long-context optimization instabilities by distilling interleaved multimodal histories into explicit reward signals for accurate causal attribution. Furthermore, to mitigate the severe latency bottleneck caused by iterative denoising, we propose $\Delta$-Space Weight Merge (DSWM), a theoretically grounded method that fuses alignment weights with off-the-shelf distillation priors, reducing the per-step inference cost to just 4 NFEs without requiring expensive re-distillation. Extensive experiments demonstrate that CLVR outperforms existing open-source baselines across multiple benchmarks and approaches the performance of proprietary commercial models, unlocking general test-time scaling capabilities for complex visual generation.
Abstract (translated)
URL
https://arxiv.org/abs/2605.14876
PDF
-
LPH-VTON: Resolving the Structure-Texture Dilemma of Virtual Try-On via Latent Process Handover
2026-05-14 14:18:56Yixin Liu, Baihong Qian, Jinglin Jiang, Jeffery Wu, Yan Chen, Wei Wang, Yida Wang, Lanqing Yang, Guangtao XueAbstract
Virtual Try-On (VTON) aims to synthesize photorealistic images of garments precisely aligned with a person's body and pose. Current diffusion-based methods, however, face a fundamental trade-off between structural integrity and textural fidelity. In this paper, we formalize this challenge as a consequence of complementary inductive biases inherent in prevailing architectures: models heavily reliant on spatial constraints naturally favor geometric alignment but often suppress textures, whereas models dominated by unconstrained generative priors excel at vibrant detail rendering but are prone to structural drift. Based on this diagnosis, we propose LPH-VTON, a new synergistic framework that resolves this tension within a single, continuous denoising process. LPH-VTON strategically decomposes the generation, leveraging a structure-biased model to establish a geometrically consistent latent scaffold in the early stages, before handing over control to a texture-biased model for high-fidelity detail rendering. Extensive experiments validate our approach. Our model achieves a superior Pareto-optimal balance, establishing new benchmarks in perceptual faithfulness while maintaining highly competitive structural alignment across the standard dataset VITON-HD, proving the efficacy of temporal architectural decoupling.
Abstract (translated)
URL
https://arxiv.org/abs/2605.14874
PDF
-
The Velocity Deficit: Initial Energy Injection for Flow Matching
2026-05-14 13:30:07Linze Li, Zong-Wei Hong, Shen Zhang, Bo Lin, Jinglun Li, Yao Tang, Jiajun LiangAbstract
While Flow Matching theoretically guarantees constant-velocity trajectories, we identify a critical breakdown in high-dimensional practice: the Velocity Deficit. We show that the MSE objective systematically underestimates velocity magnitude, causing generated samples to fail to reach the data manifold-a phenomenon we term Integration Lag. To rectify this, we propose Initial Energy Injection, instantiated via two complementary methods: the training-based Magnitude-Aware Flow Matching (MAFM) and the training-free Scale Schedule Corrector (SSC). Both are grounded in our discovery of a crucial asymmetry: velocity contraction causes harmful kinetic stagnation at the trajectory's start, yet acts as a beneficial denoising mechanism at its end. Empirically, SSC yields significant efficiency gains with zero retraining and just one line of code. On ImageNet-1k (256x256), it improves FID by 44.6% (from 13.68 to 7.58) and achieves a 5x speedup, enabling a 50-step generator (FID 7.58) to beat a 250-step baseline (FID 8.65). Furthermore, our methods generalize to Text-to-Image tasks and high-resolution generation, improving FID on MS-COCO by ~22%.
Abstract (translated)
URL
https://arxiv.org/abs/2605.14819
PDF
-
Probing into Camera Control of Video Models
Abstract
Video is a rich and scalable source of 3D/4D visual observations, and camera control is a key capability for video generation models to produce geometrically meaningful content. Existing approaches typically learn a mapping from camera motion to video using additional camera modules and paired data. However, such datasets are often limited in scale, diversity, and scene dynamics, which can bias the model toward a narrow output distribution and compromise the strong prior learned by the base model. These limitations motivate a different perspective on camera control. In this paper, we show that camera control need not be modeled as an implicit mapping problem, but can instead be treated as a form of geometric guidance that induces displacements across frames. Specifically, we reformulate camera control into a set of displacement fields and apply them via differentiable resampling of latent features during denoising. Our simple approach achieves effective camera control with minimal degradation across diverse quality metrics compared to fine-tuned baselines. Since our method is applicable to most video diffusion models without training, it can also serve as a probe to study the camera control capabilities of base models. Using this probe, we identify universal biases shared by representative video models, as well as disparities in their responses to camera control. Finally, we benchmark their performance in multi-view generation, offering insights into their potential for 3D/4D tasks.
Abstract (translated)
URL
https://arxiv.org/abs/2605.14815
PDF
-
Uncertainty Quantification for Large Language Diffusion Models
2026-05-14 08:39:56Artem Vazhentsev, Vladislav Smirnov, David Li, Maxim Panov, Timothy Baldwin, Artem ShelmanovAbstract
Large Language Diffusion Models (LLDMs) are emerging as an alternative to autoregressive models, offering faster inference through higher parallelism. Similar to autoregressive LLMs, they remain prone to hallucinations, making reliable uncertainty quantification (UQ) crucial for safe deployment. However, existing UQ methods are fundamentally misaligned with this new paradigm: they assume autoregressive factorization or use expensive repeated sampling, negating the efficiency of LLDMs. In this work, we present the first systematic study of UQ for LLDMs and propose lightweight, zero-shot uncertainty signals derived from the iterative denoising process, leveraging intermediate generations, token remasking dynamics, and denoising complexity. We further adapt a state-of-the-art UQ method to LLDMs by combining masked diffusion likelihoods with trajectory-based semantic dissimilarity. We prove that expected trajectory dissimilarity lower bounds the masked diffusion training objective, which motivates its usage as an uncertainty score. Comprehensive experiments across three tasks, eight datasets, and two models show that our method achieves a great cost-performance trade-off: it approaches the strongest sampling-based baselines while incurring up to 100x lower computational overhead. Our work demonstrates that LLDMs can deliver both fast inference and reliable hallucination detection simultaneously.
Abstract (translated)
URL
https://arxiv.org/abs/2605.14570
PDF
-
From Table to Cell: Attention for Better Reasoning with TABALIGN
2026-05-14 07:00:26Tung Sum Thomas Kwok, Zeyong Zhang, Xinyu Wang, Chunhe Wang, Xiaofeng Lin, Hanwei Wu, Lei Ding, Guang Cheng, Zhijiang GuoAbstract
Multi-step LLM reasoning over structured tables fails because planning and execution share no explicit cell-grounding contract. Existing methods constrain the planner to a left-to-right factorization at odds with table permutation invariance, and score intermediate states by generated content alone, overlooking cell grounding. We conduct a pilot study showing that diffusion language models (DLMs) produce more human-aligned and permutation-stable cell attention on tables than autoregressive models, with a 40.2% median reduction in attention-AUROC variability under row reordering. Motivated by this, we propose TABALIGN, a planned table reasoning framework that operationalizes the contract. TABALIGN pairs a masked DLM planner, whose bidirectional denoising emits plan steps as binary cell masks, with TABATTN, a lightweight verifier trained on 1,600 human-verified attention standards to score each step by its attention overlap with the plan-designated mask. Across eight benchmarks covering table question answering and fact verification, TABALIGN improves average accuracy by 15.76 percentage points over the strongest open-source baseline at comparable 8B-class scale, with a matched-backbone ablation attributing 2.87 percentage points of this gain to the DLM planner over an AR planner on a fixed reasoner. Cleaner DLM plans also accelerate downstream reasoning execution by 44.64%.
Abstract (translated)
URL
https://arxiv.org/abs/2605.14465
PDF
-
ClickRemoval: An Interactive Open-Source Tool for Object Removal in Diffusion Models
2026-05-14 06:56:53Ledun Zhang, Yatu Ji, Xufei Zhuang, Xinying YaoAbstract
Existing object removal tools often rely on manual masks or text prompts, making precise removal difficult for non-expert users in complex scenes and often leading to incomplete removal or unnatural background completion. To address this issue, we present ClickRemoval, an open-source interactive object removal tool built on pretrained Stable Diffusion models and driven solely by user clicks. Without additional training, hand-drawn masks, or text descriptions, ClickRemoval localizes target objects and restores the background through self-attention modulation during denoising. Experiments show that ClickRemoval achieves competitive results across quantitative metrics and user studies. We release a complete software package at this https URL under the Apache-2.0 license.
Abstract (translated)
URL
https://arxiv.org/abs/2605.14461
PDF
-
Where Should Diffusion Enter a Language Model? Geometry-Guided Hidden-State Replacement
2026-05-14 04:47:54Injin Kong, Hyoungjoon Lee, Yohan JoAbstract
Continuous diffusion language models lag behind autoregressive transformers, partly because diffusion is applied in spaces poorly suited to language denoising and token recovery. We propose DiHAL, a geometry-guided diffusion-transformer hybrid that asks where diffusion should enter a pretrained transformer. DiHAL scores layers with geometry-based proxies, selects a diffusion-friendly hidden-state interface, and replaces the lower transformer prefix with a diffusion bridge while retaining the upper layers and original LM head. By reconstructing the selected-layer hidden state rather than tokens, DiHAL avoids direct continuous-to-discrete recovery. Experiments on 8B-scale backbones show that the geometry score predicts effective shallow insertion layers under a fixed bridge-training protocol and that hidden-state recovery improves over continuous diffusion baselines in a diagnostic comparison matching the diffusion/recovery training budget. These results suggest that hidden-state geometry helps identify where diffusion-based replacement is feasible inside pretrained language models.
Abstract (translated)
URL
https://arxiv.org/abs/2605.14368
PDF
-
AnyBand-Diff: A Unified Remote Sensing Image Generation and Band Repair Framework with Spectral Priors
2026-05-14 04:04:34Zuopeng Zhao, Ying Liu, Xiaoyu Li, Su Luo, Lu Li, Wenwen LiuAbstract
Existing diffusion models have made significant progress in generating realistic images. However, their direct adaptation to remote sensing imagery often disregards intrinsic physical laws. This oversight frequently leads to spectral distortion and radiometric inconsistency, severely limiting the scientific utility of generated data. To address this issue, this paper introduces AnyBand-Diff, a novel spectral-prior-guided diffusion framework tailored for robust spectral reconstruction. Specifically, we design a Masked Conditional Diffusion backbone integrated with a dual stochastic masking strategy, empowering the model to recover complete spectral information from arbitrary band subsets. Subsequently, to ensure radiometric fidelity, a Physics-Guided Sampling mechanism is proposed, leveraging gradients from a differentiable physical model to explicitly steer the denoising trajectory toward the manifold of physically plausible solutions. Furthermore, a Multi-Scale Physical Loss is formulated to enforce rigorous constraints across pixel, region, and global levels in a joint manner. Extensive experiments confirm the effectiveness of AnyBand-Diff in generating reliable imagery and achieving accurate spectral reconstruction, contributing to the advancement of physics-aware generative methods for Earth observation.
Abstract (translated)
URL
https://arxiv.org/abs/2605.14341
PDF
-
Factorization-Error-Free Discrete Diffusion Language Model via Speculative Decoding
2026-05-14 03:15:25Xun Fang, Yunchen Li, Hang Yuan, Zhou YuAbstract
Discrete diffusion language models improve generation efficiency through parallel token prediction, but standard $X_0$ prediction methods introduce factorization errors by approximating the clean token posterior with independent token-wise distributions. This paper proposes Factorization-Error-Free Discrete Diffusion Language Modeling (FeF-DLLM), which replaces independent clean-token prediction with an exact prefix-conditioned factorization of the clean posterior to better preserve token dependencies. To reduce the sequential cost introduced by prefix conditioning, FeF-DLLM further incorporates speculative decoding within diffusion denoising, accelerating inference while maintaining the parallel prediction and re-masking properties of DLLMs. Theoretically, we prove that FeF-DLLM generates from the true joint distribution and derive its expected acceleration ratio. Experiments on GSM8K, MATH, HumanEval, and MBPP demonstrate that our method improves accuracy by an average of 5.04 percentage points while achieving an average inference speedup of $3.86\times$.
Abstract (translated)
URL
https://arxiv.org/abs/2605.14305
PDF
-
CoReDiT: Spatial Coherence-Guided Token Pruning and Reconstruction for Efficient Diffusion Transformers
2026-05-13 23:13:29Zhuojin Li, Hsin-Pai Cheng, Hong Cai, Shizhong Han, Fatih PorikliAbstract
Diffusion Transformers (DiTs) deliver remarkable image and video generation quality but incur high computational cost, limiting scalability and on-device deployment. We introduce CoReDiT, a structured token pruning framework for DiTs across vision tasks. CoReDiT uses a linear-time spatial coherence score to estimate local redundancy in the latent token lattice and skips high coherence (redundant) tokens in self-attention. To maintain a dense representation and avoid visual discontinuities, we reconstruct skipped attention outputs via coherence-guided aggregation of spatially neighboring retained tokens. We further introduce a progressive, block-adaptive pruning schedule that increases pruning gradually and allocates larger budgets to blocks and denoising steps with higher redundancy. Across state-of-the-art diffusion backbones including PixArt-{\alpha} and MagicDrive-V2, CoReDiT achieves up to 55% self-attention FLOPs reduction and inference speedups of 1.33x on cloud GPUs and 1.72x on mobile NPUs, while maintaining high visual quality. Notably, CoReDiT also increases on-device memory head-room, enabling higher-resolution generation.
Abstract (translated)
URL
https://arxiv.org/abs/2605.14191
PDF
-
OmniLiDAR: A Unified Diffusion Framework for Multi-Domain 3D LiDAR Generation
2026-05-13 17:42:20Youquan Liu, Weidong Yang, Ao Liang, Xiang Xu, Lingdong Kong, Yang Wu, Dekai Zhu, Xin Li, Runnan Chen, Ben Fei, Tongliang Liu, Wanli OuyangAbstract
LiDAR scene generation is increasingly important for scalable simulation and synthetic data creation, especially under diverse sensing conditions that are costly to capture at scale. Typically, diffusion-based LiDAR generators are developed under single-domain settings, requiring separate models for different datasets or sensing conditions and hindering unified, controllable synthesis under heterogeneous distribution shifts. To this end, we present OmniLiDAR, a unified text-conditioned diffusion framework that generates LiDAR scans in a shared range-image representation across eight representative domains spanning three shift types: adverse weather, sensor-configuration changes (e.g., reduced beams), and cross-platform acquisition (vehicle, drone, and quadruped). To enable training a single model over heterogeneous domains without isolating optimization by domain, we introduce a Cross-Domain Training Strategy (CDTS) that mixes domains within each mini-batch and leverages conditioning to steer generation. We further propose Cross-Domain Feature Modeling (CDFM), which captures directional dependencies along azimuth and elevation axes to reflect the anisotropic scanning structure of range images, and Domain-Adaptive Feature Scaling (DAFS) as a lightweight modulation to account for structured domain-dependent feature shifts during denoising. In the absence of a public consolidated benchmark, we construct an 8-domain dataset by combining real-world scans with physically based weather simulation and systematic beam reduction while following official splits. Extensive experiments demonstrate strong generation fidelity and consistent gains in downstream use cases, including generative data augmentation for LiDAR semantic segmentation and 3D object detection, as well as robustness evaluation under corruptions, with consistent benefits in limited-label regimes.
Abstract (translated)
URL
https://arxiv.org/abs/2605.13815
PDF
