| Total: 3352
Recent advances in multimodal large language models (MLLMs) have substantially expanded the capabilities of multimodal retrieval, enabling systems to align and retrieve information across visual and textual modalities. Yet, existing benchmarks largely focus on coarse-grained or single-condition alignment, overlooking real-world scenarios where user queries specify multiple interdependent constraints across modalities. To bridge this gap, we introduce MCMR (Multi-Conditional Multimodal Retrieval): a large-scale benchmark designed to evaluate fine-grained, multi-condition cross-modal retrieval under natural-language queries. MCMR spans five product domains: upper and bottom clothing, jewelry, shoes, and furniture. It also preserves rich long-form metadata essential for compositional matching. Each query integrates complementary visual and textual attributes, requiring models to jointly satisfy all specified conditions for relevance. We benchmark a diverse suite of MLLM-based multimodal retrievers and vision-language rerankers to assess their condition-aware reasoning abilities. Experimental results reveal: (i) distinct modality asymmetries across models; (ii) visual cues dominate early-rank precision, while textual metadata stabilizes long-tail ordering; and (iii) MLLM-based pointwise rerankers markedly improve fine-grained matching by explicitly verifying query-candidate consistency. Overall, MCMR establishes a challenging and diagnostic benchmark for advancing multimodal retrieval toward compositional, constraint-aware, and interpretable understanding. Our code and dataset is available at https://github.com/EIT-NLP/MCMR
The performance of multimodal learning systems, particularly in high-stakes domains like automated depression recognition, is fundamentally constrained by the challenge of learning robust visual representations from limited and complex clinical data. To overcome this, we introduce Cross-Modal Guided Visual Synthesis (CMG-VS), a novel training framework that internally enhances the learning process by synthesizing new, task-relevant visual features. At its core, CMG-VS leverages the rich context from audio and text modalities to guide a conditional generative model. This model learns the intricate mapping from speech and language to visual expression, generating a diverse manifold of plausible visual behaviors to enrich the training distribution. Crucially, this synthesis is not a separate pre-processing step. Through a task-guided joint optimization scheme, the generative process is dynamically steered by the downstream multimodal recognizer's performance. This closed-loop feedback mechanism ensures the synthesized visual features are optimized to be maximally discriminative for the recognition task, rather than merely realistic. Comprehensive experiments on the widely-used DAIC-WOZ and E-DAIC benchmark datasets demonstrate that CMG-VS significantly outperforms existing state-of-the-art methods across all standard regression and classification metrics. Ablation studies further validate that our task-guided synthesis is the key driver of this performance gain, proving its effectiveness as a new paradigm for robust multimodal representation learning.
We present LoD-Loc v3, a novel method for generalized aerial visual localization in dense urban environments. While prior work LoD-Loc v2 achieves localization through semantic building silhouette alignment with low-detail city models, it suffers from two key limitations: poor cross-scene generalization and frequent failure in dense building scenes. Our method addresses these challenges through two key innovations. First, we develop a new synthetic data generation pipeline that produces InsLoD-Loc - the largest instance segmentation dataset for aerial imagery to date, comprising 100k images with precise instance building annotations. This enables trained models to exhibit remarkable zero-shot generalization capability. Second, we reformulate the localization paradigm by shifting from semantic to instance silhouette alignment, which significantly reduces pose estimation ambiguity in dense scenes. Extensive experiments demonstrate that LoD-Loc v3 outperforms existing state-of-the-art (SOTA) baselines, achieving superior performance in both cross-scene and dense urban scenarios with a large margin. The project is available at https://nudt-sawlab.github.io/LoD-Locv3/.
Vision-language models (VLMs) have achieved impressive results on single-view vision tasks, but lack the multi-view spatial reasoning capabilities essential for embodied AI systems to understand 3D environments and manipulate objects across different viewpoints. In this work, we introduce Cross-View Relations (XVR), a large-scale dataset designed to teach VLMs spatial reasoning across multiple views. XVR comprises 100K vision-question-answer samples derived from 18K diverse 3D scenes and 70K robotic manipulation trajectories, spanning three fundamental spatial reasoning tasks: Correspondence (matching objects across views), Verification (validating spatial relationships), and Localization (identifying object positions). VLMs fine-tuned on XVR achieve substantial improvements on established multi-view and robotic spatial reasoning benchmarks (MindCube and RoboSpatial). When integrated as backbones in Vision-Language-Action models, XVR-trained representations improve success rates on RoboCasa. Our results demonstrate that explicit training on cross-view spatial relations significantly enhances multi-view reasoning and transfers effectively to real-world robotic manipulation.
3D Gaussian Splatting reconstructs scenes by starting from a sparse Structure-from-Motion initialization and refiningunder-reconstructed regions. This process is slow, as it requires multiple densification steps where Gaussians arerepeatedly split and adjusted, following a lengthy optimization path. Moreover, this incremental approach often yieldssuboptimal renderings in high-frequency regions. We propose a fundamentally different approach: eliminate densification with a one-step approximation of scenegeometry using triangulated pixels from dense image correspondences. This dense initialization allows us to estimatethe rough geometry of the scene while preserving rich details from input RGB images, providing each Gaussian withwell-informed color, scale, and position. As a result, we dramatically shorten the optimization path and remove theneed for densification. Unlike methods that rely on sparse keypoints, our dense initialization ensures uniform detailacross the scene, even in high-frequency regions where other methods struggle. Moreover, since all splats are initializedin parallel at the start of optimization, we remove the need to wait for densification to adjust new Gaussians.EDGS reaches LPIPS and SSIM performance of standard 3DGS significantly faster than existing efficiency-focused approaches. When trained further, it exceeds the reconstruction quality of state-of-the-art models aimed at maximizing fidelity. Our method is fully compatible with other acceleration techniques, making it a versatile and efficient solution that can be integrated with existing approaches.
Video Unsupervised Domain Adaptation (VUDA) poses a significant challenge in action recognition, requiring the adaptation of a model from a labeled source domain to an unlabeled target domain. Despite recent advances, existing VUDA methods often fall short of fully supervised performance, a key reason being the prevalence of static and uninformative backgrounds that exacerbate domain shifts. Additionally, prior approaches largely overlook computational efficiency, limiting real-world adoption. To address these issues, we propose Learnable Motion-Focused Tokenization (LMFT) for VUDA. LMFT tokenizes video frames into patch tokens and learns to discard low-motion, redundant tokens, primarily corresponding to background regions, while retaining motion-rich, action-relevant tokens for adaptation. Extensive experiments on three standard VUDA benchmarks across 21 domain adaptation settings show that our VUDA framework with LMFT achieves state-of-the-art performance while significantly reducing computational overhead. LMFT thus enables VUDA that is both effective and computationally efficient.
Vision-Language-Action (VLA) models are advancing autonomous driving by replacing modular pipelines with unified end-to-end architectures. Current VLAs face two challenges: (1) they require extensive datasets annotated with reasoning traces, and (2) these traces greatly increase token counts, inflating training and inference costs. We propose NoRD (No Reasoning for Driving), a data- and inference-efficient VLA that addresses both. Compared to existing VLAs, NoRD achieves competitive performance while being fine-tuned on atleast <60% of the data and no reasoning annotations, resulting in 3x fewer tokens. Our approach applies Reinforcement Learning (RL) to fine-tune a Supervised fine-tuning (SFT) policy trained on a small, reasoning-free dataset. However, we observe that the standard RL algorithm, Group Relative Policy Optimization (GRPO), fails to yield significant improvements over this data-efficient SFT policy. We find that this limitation stems from difficulty bias, which disproportionately penalizes reward signals from scenarios that produce high-variance rollouts within GRPO. NoRD overcomes this limitation by incorporating Dr.GRPO, a recent algorithm designed to mitigate difficulty bias in LLMs. As a result, NoRD achieves competitive performance on Waymo and NAVSIM without large datasets, reasoning or additional inputs, enabling scalable, data-efficient training, and fast inference.
While the most fundamental pretraining paradigm typically trains modality-specific models on their respective datasets, the Platonic Representation Hypothesis that representations eventually align across modalities as data and model scale suggests an intriguing possibility: large language models (LLMs) could be pretrained on visual corpora to reach parity with text-pretrained models, thereby expanding data sources to break the text-scaling bottlenecks, and leveraging richer visual cues for more comprehensive corpus understanding. This paper makes the first attempt to demonstrate the feasibility of this implication by introducing Masked Autoregressive Pretraining for Learning language intelligencE (MAPLE), a novel visual pretraining paradigm for LLMs that leverages raw document images to improve language intelligence. MAPLE is universal to integrate masked auto-regressive models with various LLM backbones, where the LLMs are incentivized to generate latent hypotheses for the masked regions based on the unmasked regions. We verify MAPLE in the domain of math reasoning with multiple LLM backbones and show that MAPLE consistently surpasses text-only pretraining relatively by at most 40.2% on average accuracy across four math reasoning benchmarks. Further analyses show that visually pretrained LLMs learn a shared latent space that aligns document visuals with text and exploits layout and structural cues, supporting visual pretraining as a feasible and scalable route to stronger language models.
Temporal information is crucial for visual tracking, but existing multi-frame trackers are vulnerable to model drift caused by naively aggregating noisy historical predictions. In this paper, we introduce DTPTrack, a lightweight and generalizable module designed to be seamlessly integrated into existing trackers to suppress drift. Our framework consists of two core components: (1) a Temporal Reliability Calibrator (TRC) mechanism that learns to assign a per-frame reliability score to historical states, filtering out noise while anchoring on the ground-truth template; and (2) a Temporal Guidance Synthesizer (TGS) module that synthesizes this calibrated history into a compact set of dynamic temporal priors to provide predictive guidance. To demonstrate its versatility, we integrate DTPTrack into three diverse tracking architectures--OSTrack, ODTrack, and LoRAT--and show consistent, significant performance gains across all baselines. Our best-performing model, built upon an extended LoRATv2 backbone, sets a new state-of-the-art on several benchmarks, achieving a 77.5% Success rate on LaSOT and an 80.3% AO on GOT-10k. The source code is available at https://github.com/NorahGreen/DTPTrack.
Graphical User Interface (GUI) agents have the potential to assist users in interacting with complex software (e.g., PowerPoint, Photoshop). While prior research has primarily focused on automating user actions through clicks and keystrokes, this paradigm overlooks human intention, where users value the ability to explore, iterate, and refine their ideas while maintaining agency. To move beyond automation and toward collaboration, GUI agents must understand what users are doing and why. We introduce GUIDE (GUI User Intent Detection Evaluation), a benchmark that evaluates AI models on their ability to perceive user behavior, infer intent, and provide assistance in open-ended GUI tasks. GUIDE consists of 67.5 hours of screen recordings from 120 novice user demonstrations with think-aloud narrations, across 10 software. GUIDE defines three tasks--(i) Behavior State Detection, (ii) Intent Prediction, and (iii) Help Prediction that test a model's ability to recognize behavior state, reason about goals, and decide when and how to help. Evaluations across eight state-of-the-art multimodal models reveal that all models struggled, achieving only 44.6% and 55.0% accuracy on behavior state and help prediction. However, providing user context significantly improved the performance, raising help prediction by up to 50.2pp, highlighting the critical role of structured user understanding in effective assistance. Our dataset is available at https://guide-bench.github.io.
Multimodal Deepfakes proliferating on social media threaten authenticity, information integrity, and digital forensics. Existing benchmarks are constrained by their single-modality scope, simplified manipulations, or unrealistic distributions, which limit their ability to assess real-world robustness. We present Omni-Fake, a unified omni-dataset for comprehensive multimodal deepfake detection in social-media settings. It comprises Omni-Fake-Set, a large-scale, high-quality dataset with 1M+ samples, and Omni-Fake-OOD, an out-of-distribution benchmark with 100k+ samples intentionally excluded from training to evaluate generalization. Omni-Fake spans four modalities--image, audio, video, and audio-video talking head and supports a joint detection-localization-explanation protocol. For images, audio, and videos, we define a ternary task (real / partially manipulated / fully synthetic) with spatial or temporal localization masks for fine-grained reasoning. Talking heads are formulated as an audio-video fusion binary task targeting speaking digital humans and lip-synced avatar forgeries. On top of Omni-Fake, we further propose Omni-Fake-R1, a reinforcement-learning-driven multimodal detector that adaptively integrates visual and auditory cues and outputs structured decisions, localization, and natural-language explanations. Extensive experiments show significant gains in detection accuracy, cross-modal generalization, and explainability over state-of-the-art baselines. Code will be released.
Humanoid robotics has strong potential to transform daily service and caregiving applications. Although recent advances in general motion tracking within physics engines (GMT) have enabled virtual characters and humanoid robots to reproduce a broad range of human motions, these behaviors are primarily limited to contact-less social interactions or isolated movements. Assistive scenarios, by contrast, require continuous awareness of a human partner and rapid adaptation to their evolving posture and dynamics. In this paper, we formulate the imitation of closely interacting, force-exchanging human-human motion sequences as a multi-agent reinforcement learning problem. We jointly train partner-aware policies for both the supporter (assistant) agent and the recipient agent in a physics simulator to track assistive motion references. To make this problem tractable, we introduce a partner policies initialization scheme that transfers priors from single-human motion-tracking controllers, greatly improving exploration. We further propose dynamic reference retargeting and contact-promoting reward, which adapt the assistant's reference motion to the recipient's real-time pose and encourage physically meaningful support. We show that \model is the first method capable of successfully tracking assistive interaction motions on established benchmarks, demonstrating the benefits of a multi-agent RL formulation for physically grounded and socially aware humanoid control.
Prospective reconstruction is crucial in many clinical applications such as MRI-guided radiotherapy, which demands accurate image reconstruction and fast motion estimation from currently acquired measurements. However, prospective reconstruction remains challenging due to ultra-sparse sampling and stringent latency requirements. In this work, we propose PDMR, an Prospective Dynamic 3D MRI Reconstruction framework with latent-space motion tracking. Our core idea is to learn an efficient and generalizable latent manifold of motion fields offline, enabling rapid online adaptation for prospective reconstruction. Specifically, we parameterize the deformation vector fields (DVFs) on a low-dimensional manifold, effectively reducing the search space for fast online adaptation, and employ a tri-plane representation to achieve geometry-aware and memory-efficient encoding of 3D motion. Experiments on both XCAT digital phantoms and in-house abdominal MRI datasets demonstrate that PDMR achieves high-fidelity and temporally consistent reconstruction across multiple prospective scenarios (Immediate and After-2min), outperforming state-of-the-art retrospective and online methods. Our results suggest a promising pathway toward ultra-fast, motion-aware prospective MRI reconstruction in clinical practice.
Generating dynamic and interactive 3D trees has wide applications in virtual reality, games, and world simulation. However, existing methods still face various challenges in generating structurally consistent and realistic 4D motion for complex real trees. In this paper, we propose DynamicTree, the first framework that can generate long-term, interactive 3D motion for 3DGS reconstructions of real trees. Unlike prior optimization-based methods, our approach generates dynamics in a fast feed-forward manner. The key success of our approach is the use of a compact sparse voxel spectrum to represent the tree movement. Given a 3D tree from Gaussian Splatting reconstruction, our pipeline first generates mesh motion using the sparse voxel spectrum and then binds Gaussians to deform the mesh. Additionally, the proposed sparse voxel spectrum can also serve as a basis for fast modal analysis under external forces, allowing real-time interactive responses. To train our model, we also introduce 4DTree, the first large-scale synthetic 4D tree dataset containing about 8.5k animated tree meshes with semantic labels and 100-frame motion sequences. Extensive experiments demonstrate that our method achieves realistic and responsive tree animations, significantly outperforming existing approaches in both visual quality and computational efficiency.
Gait recognition has emerged as a powerful biometric technique for identifying individuals at a distance without requiring user cooperation. Most existing methods focus primarily on RGB-derived modalities, which fall short in real-world scenarios requiring multi-modal collaboration and cross-modal retrieval. To overcome these challenges, we present MMGait, a comprehensive multi-modal gait benchmark integrating data from five heterogeneous sensors, including an RGB camera, a depth camera, an infrared camera, a LiDAR scanner, and a 4D Radar system. MMGait contains twelve modalities and 334,060 sequences from 725 subjects, enabling systematic exploration across geometric, photometric, and motion domains. Based on MMGait, we conduct extensive evaluations on single-modal, cross-modal, and multi-modal paradigms to analyze modality robustness and complementarity. Furthermore, we introduce a new task, Omni Multi-Modal Gait Recognition, which aims to unify the above three gait recognition paradigms within a single model. We also propose a simple yet powerful baseline, OmniGait, which learns a shared embedding space across diverse modalities and achieves promising recognition performance. The MMGait benchmark, codebase, and pretrained checkpoints are publicly available at https://github.com/BNU-IVC/MMGait.
Inspired by the success of Reinforcement Learning with Human Feedback (RLHF) in image generation, recent work has adapted reward-based learning to image super-resolution (ISR) by using Image Quality Assessment (IQA) models as rewards. However, existing IQA models typically output only a single global score and are insensitive to local, fine-grained distortions, allowing perceptually undesirable artifacts to obtain spuriously high rewards and leading to reward hacking. To address this issue, we propose FinPercep-RM, a fine-grained perceptual reward model built on an encoder-decoder architecture that predicts both a global quality score and a Perceptual Degradation Map for spatially localizing and quantifying local defects. We further introduce FGR-30k, a dataset containing diverse and subtle distortions produced by real-world super-resolution models, to train the reward model. While FinPercep-RM provides stronger supervision, its increased complexity also makes generator policy learning unstable. We therefore develop a Co-evolutionary Curriculum Learning (CCL) strategy, in which the reward model and the ISR model evolve synchronously: the reward signal progressively increases in complexity, while the ISR model starts with simple global supervision for fast convergence and gradually transitions to fine-grained rewards. This easy-to-hard design stabilizes training and suppresses reward hacking. Extensive experiments across multiple ISR models demonstrate improvements in both global quality and local realism. Code will be available at https://github.com/lyd-2022/FinPercep-RM.
Continual learning for video--language understanding is increasingly important as models face non-stationary data, domains, and query styles, yet prevailing solutions blur what should stay stable versus what should adapt, rely on static routing/capacity, or require replaying past videos. We aim to explicitly specify where stability lives and where plasticity should be focused under realistic memory and privacy constraints. We introduce Affordance-First Decomposition (AFD): videos are mapped to slowly varying affordance tokens that form a shared, time-aligned substrate, while a lightweight, query-routed, conflict-aware scheduler concentrates adaptation and grows capacity only when needed. The substrate is stabilized via weak alignment and teacher consistency, and training uses question-only replay. AFD achieves state-of-the-art across protocols: 51.6% average accuracy with -1.8% forgetting on domain-incremental VideoQA, ViLCo R@1@0.5 of 29.6% (MQ) and 20.7% (NLQ) with 18.4% stAP@0.25 (VQ), and 39.5% accuracy with -1.6% forgetting on time-incremental iVQA. Overall, AFD offers an explicit, interpretable split between a stable interaction-centered substrate and targeted adaptation.
Diffusion models have recently motivated great success in many generation tasks like object removal. Nevertheless, existing image decomposition methods struggle to disentangle semi-transparent or transparent layer occlusions due to mask prior dependencies, static object assumptions, and the lack of datasets. In this paper, we delve into a novel task: Layer-Wise Decomposition of Alpha-Composited Images, aiming to recover constituent layers from single overlapped images under the condition of semi-transparent/transparent alpha layer non-linear occlusion. To address challenges in layer ambiguity, generalization, and data scarcity, we first introduce AlphaBlend, the first large-scale and high-quality dataset for transparent and semi-transparent layer decomposition, containing six subtasks with different characteristics (e.g., translucent flare removal, semi-transparent cell decomposition, glassware decomposition). Building on this dataset, we present DiffDecompose, a diffusion Transformer-based framework that learns the posterior over possible layer decompositions conditioned on the input image, semantic prompts, and blending type. Rather than regressing alpha mattes directly, DiffDecompose performs In-Context Decomposition, enabling the model to predict one or multiple layers without per-layer supervision, and introduces Layer Position Encoding Cloning to maintain pixel-level correspondence across layers. Extensive experiments on the proposed AlphaBlend dataset and public LOGO dataset verify the effectiveness of DiffDecompose. Code will be publicly available at https://github.com/Wangzt1121/DiffDecompose.
Color transfer aims to match the color distribution of a content image (source) to that of a style image (target) while preserving structure and perceptual realism. Yet modulation-based flow models such as ModFlows often produce trajectory misalignment and artifacts because they rely on strictly linear transport paths. We propose NCT, a nonlinear color transfer framework that replaces linear paths with Bezier trajectories, enabling smooth, nonlinear, and perceptually coherent color transfer. This parameterization lets the transport bend toward plausible intermediate color regimes, improving content-style alignment and reducing chromatic distortion. We further incorporate a Mixture of Experts (MoE) module in the encoder to select trajectory experts for different chromatic regimes, improving generalization to heterogeneous data with complex illumination and materials. Experiments show that NCT reduces artifacts and achieves more stable color transfer than prior flow-based methods, especially on 3D-rendered or highly textured images. The code is provided in supplementary materials.
While diffusion models have achieved great success in the field of video generation, this progress is accompanied by a rapidly escalating computational burden. Among the existing acceleration methods, Feature Caching is popular due to its training-free property and considerable speedup performance, but it inevitably faces semantic and detail drop with further compression. Another widely adopted method, training-aware step-distillation, though successful in image generation, also faces drastic degradation in video generation with a few steps. Furthermore, the quality loss becomes more severe when simply applying training-free feature caching to the step-distilled models, due to the sparser sampling steps. This paper novelly introduces a distillation-compatible learnable feature caching mechanism for the first time. We employ a lightweight learnable neural predictor instead of traditional training-free heuristics for diffusion models, enabling a more accurate capture of the high-dimensional feature evolution process. Furthermore, we explore the challenges of highly compressed distillation on large-scale video models and propose a conservative Restricted MeanFlow approach to achieve more stable and lossless distillation. By undertaking these initiatives, we further push the acceleration boundaries to 11.8 times while preserving generation quality. Extensive experiments demonstrate the effectiveness of our method.
Despite major advances brought by diffusion-based models, current 3D texture generation systems remain hindered by cross-view inconsistency -- textures that appear convincing from one viewpoint often fail to align across others. We find that this issue arises from attention ambiguity, where unstructured full attention is applied indiscriminately across tokens and modalities, causing geometric confusion and unstable appearance-structure coupling.To address this, we introduce CaliTex, a framework of geometry-calibrated attention that explicitly aligns attention with 3D structure.It introduces two modules: Part-Aligned Attention that enforces spatial alignment across semantically matched parts, and Condition-Routed Attention which routes appearance information through geometry-conditioned pathways to maintain spatial fidelity.Coupled with a two-stage diffusion transformer, CaliTex makes geometric coherence an inherent behavior of the network rather than a byproduct of optimization.Empirically, CaliTex produces seamless and view-consistent textures and outperforms both open-source and commercial baselines.
Aerial-ground visual localization is a challenging task due to the significant differences in scene scale and view point captured between two views. In this work, we explore the practical benefit of jointly learning camera calibration and bird's-eye-view (BEV) projection for estimating full 6 Degrees-of-freedom relative camera pose between uncalibrated aerial and ground views. We present Visual Geometry Alignment (VGA), a unified framework that jointly learns a global gravity-alignment prior inferred from dense monocular perspective fields, and a planar alignment prior complementing the unobserved azimuth angle through Procrustes alignment in a shared BEV plane. At inference, we jointly refine the relative camera pose by integrating the predicted per-camera gravity alignment and relative planar azimuth angle, yielding improved orientation and translation alignment from visual input with extreme wide base-lines and limited overlap. We evaluate our method on challenging MatrixCity, ACC-NVS1 and ULTRRA ground-aerial pairs, demonstrating that optimizing with learned geometric priors can further improve the camera pose estimation across diverse altitudes and environment.
Current multi-view indoor 3D object detectors rely on sensor geometry that is costly to obtain--i.e., precisely calibrated multi-view camera poses--to fuse multi-view information into a global scene representation, limiting deployment in real-world scenes. We target a more practical setting: Sensor-Geometry-Free (SG-Free) multi-view indoor 3D object detection, where there are no sensor-provided geometric inputs (multi-view poses or depth). Recent Visual Geometry Grounded Transformer (VGGT) shows that strong 3D cues can be inferred directly from images. Building on this insight, we present VGGT-Det, the first framework tailored for SG-Free multi-view indoor 3D object detection. Rather than merely consuming VGGT predictions, our method integrates VGGT encoder into a transformer-based pipeline. To effectively leverage both the semantic and geometric priors from inside VGGT, we introduce two novel key components: (i) Attention-Guided Query Generation (AG): exploits VGGT attention maps as semantic priors to initialize object queries, improving localization by focusing on object regions while preserving global spatial structure. (ii) Query-Driven Feature Aggregation (QD): a learnable See-Query interacts with object queries to 'see' what they need, then dynamically aggregates multi-level geometric features across VGGT layers that progressively lift 2D features into 3D. Experiments show that VGGT-Det significantly surpasses SG-Free baselines by 4.4 and 8.6 mAP@0.25 on ScanNet and ARKitScenes, respectively. Ablation study shows that VGGT's internally learned semantic and geometric priors can be effectively leveraged by our AG and QD. Source code and pre-trained models are available at the GitHub project page: https://github.com/yangcaoai/VGGT-Det-CVPR2026
Numerous techniques have been proposed for generating adversarial examples under strict Lp-norm constraints. However, such norm-bounded examples often fail to align well with human perception, and only a few methods specifically explore perceptually aligned adversarial examples. Moreover, it remains unclear whether insights from Lp-constrained attacks can be effectively leveraged to improve perceptual efficacy. In this paper, we introduce DASH, a differentiable meta-attack framework that generates effective and perceptually aligned adversarial examples by strategically composing existing Lp-based attack methods. DASH operates in a multi-stage fashion: at each stage, it aggregates candidate adversarial examples from multiple base attacks using learned, adaptive weights and propagates the result to the next stage. A meta-loss function guides this process by jointly minimizing misclassification loss and perceptual distortion, enabling the framework to dynamically modulate the contribution of each base attack throughout the stages. We evaluate DASH on adversarially trained robust models across CIFAR-10, CIFAR-100, and ImageNet while considering visual perception metrics (e.g. SSIM, FID, LPIPS) in the perturbation budget (instead of Lp-norm). Despite relying solely on Lp-constrained based methods, DASH significantly outperforms state-of-the-art perceptual attacks such as AdvAD, achieving higher attack success rates (e.g., 20.63% improvement) and superior visual quality, as measured by SSIM, LPIPS, and FID (improvements of 11, 0.015, and 5.7, respectively). DASH generalizes well to unseen defenses and different white-box/black-box scenarios, making it a practical and strong baseline for evaluating robustness.