We propose a CompliantVLA-adaptor that augments the state-of-the-art Vision-Language-Action (VLA) models with vision-language model (VLM)-informed context-aware variable impedance control (VIC) to improve the safety and effectiveness of contact-rich robotic manipulation tasks. Existing VLA systems (e.g., RDT, Pi0.5, OpenVLA-oft) typically output position, but lack force-aware adaptation, leading to unsafe or failed interactions in physical tasks involving contact, compliance, or uncertainty. In the proposed CompliantVLA-adaptor, a VLM interprets task context from images and natural language to adapt the stiffness and damping parameters of a VIC controller. These parameters are further regulated using real-time force/torque feedback to ensure interaction forces remain within safe thresholds. We demonstrate that our method outperforms the VLA baselines on a suite of complex contact-rich tasks, both in simulation and the real world, with improved success rates and reduced force violations. This work presents a promising path towards a safe foundation model for physical contact-rich manipulation. We release our code, prompts, and force-torque-impedance-scenario context datasets at https://sites.google.com/view/compliantvla.

The dynamic multi-mode resource-constrained project scheduling problem is a challenging scheduling problem that requires making decisions on both the execution order of activities and their corresponding execution modes. Genetic programming has been widely applied as a hyper-heuristic to evolve priority rules that guide the selection of activity-mode pairs from the current eligible set. Recently, an activity group selection strategy has been proposed to select a subset of activities rather than a single activity at each decision point, allowing for more effective scheduling by considering the interdependence between activities. Although effective in small-scale instances, this strategy suffers from scalability issues when applied to larger problems. In this work, we enhance the scalability of the group selection strategy by introducing a knee-point-based selection mechanism to identify a promising subset of activities before evaluating their combinations. An activity ordering rule is first used to rank all eligible activity-mode pairs, followed by a knee point selection to find the promising pairs. Then, a group selection rule selects the best activity combination. We develop a multi-tree GP framework to evolve both types of rules simultaneously. Experimental results demonstrate that our approach scales well to large instances and outperforms GP with sequential decision-making in most scenarios.

Vision-language models (VLMs) lag behind text-only language models on mathematical reasoning when the same problems are presented as images rather than text. We empirically characterize this as a modality gap: the same question in text form yields markedly higher accuracy than its visually typeset counterpart, due to compounded failures in reading dense formulas, layout, and mixed symbolic-diagrammatic context. First, we introduce VisTIRA (Vision and Tool-Integrated Reasoning Agent), a tool-integrated reasoning framework that enables structured problem solving by iteratively decomposing a given math problem (as an image) into natural language rationales and executable Python steps to determine the final answer. Second, we build a framework to measure and improve visual math reasoning: a LaTeX-based pipeline that converts chain-of-thought math corpora (e.g., NuminaMath) into challenging image counterparts, and a large set of synthetic tool-use trajectories derived from a real-world, homework-style image dataset (called SnapAsk) for fine-tuning VLMs. Our experiments show that tool-integrated supervision improves image-based reasoning, and OCR grounding can further narrow the gap for smaller models, although its benefit diminishes at scale. These findings highlight that modality gap severity inversely correlates with model size, and that structured reasoning and OCR-based grounding are complementary strategies for advancing visual mathematical reasoning.

While contemporary Vision-Language Models (VLMs) excel at 2D visual understanding, they remain constrained by a passive, 2D-centric paradigm that severely limits genuine 3D spatial reasoning. To bridge this gap, we introduce Think3D, a novel framework that equips VLM agents with interactive, 3D chain-of-thought reasoning capabilities. By integrating a suite of 3D manipulation tools, Think3D transforms passive perception into active spatial exploration, closely mirroring human geometric reasoning. We demonstrate that Think3D acts as a highly effective zero-shot plug-in for state-of-the-art closed-source models (e.g., GPT-4.1, Gemini 2.5 Pro), yielding absolute performance gains of +7.8% on BLINK Multi-view and MindCube, and +4.7% on VSI-Bench. Furthermore, to optimize tool-use in smaller open-weight models, we propose Think3D-RL, a reinforcement learning paradigm designed to autonomously learn spatial exploration strategies. When applied to Qwen3-VL-4B, Think3D-RL amplifies the performance gain from a marginal +0.7% to a substantial +10.7%. Notably, this RL formulation induces an exploration policy that qualitatively aligns with the sophisticated behavior of much larger models, entirely circumventing the need for costly operation-trajectory annotations. Ultimately, Think3D establishes tool-augmented active exploration as an effective paradigm for unlocking human-like 3D reasoning in multimodal agents. Code, models, and data are available at https://github.com/zhangzaibin/spagent.

Solar activity, including solar flares, coronal mass ejections (CMEs), and geomagnetic storms can significantly impact satellites, aviation, power grids, data centers, and space missions. Extreme solar events can cause substantial economic damage with limited advance warning, underscoring the importance of early warning systems, accurate forecasting, and effective education in space science. Although large language models (LLMs) perform well on general tasks, they often lack domain specific knowledge and pedagogical capability to clearly explain complex space science concepts. We introduce SolarGPT-QA, a question answering system based on a domain adapted large language model built on the LLaMA-3 base model. The model is trained using scientific literature and large scale question and answer data generated with GPT-4 and refined using Grok-3 in a student friendly storytelling style. To evaluate response quality, we employ an LLM-as-judge evaluation framework, where a strong reference model assesses generated answers using structured criteria including scientific accuracy, clarity, completeness, and pedagogical effectiveness. Results show that SolarGPT-QA performs strongly relative to general purpose models in zero shot settings and achieves competitive performance compared to instruction tuned models for educational explanations in space weather and heliophysics. Ablation studies indicate that combining domain adaptive pretraining with fine tuning is important for balancing scientific accuracy and educational effectiveness.

As hubs of human activity, urban surfaces consist of a wealth of semantic entities. Segmenting these various entities from satellite imagery is crucial for a range of downstream applications. Current advanced segmentation models can reliably segment entities defined by physical attributes (e.g., buildings, water bodies) but still struggle with socially defined categories (e.g., schools, parks). In this work, we achieve socio-semantic segmentation by vision-language model reasoning. To facilitate this, we introduce the Urban Socio-Semantic Segmentation dataset named SocioSeg, a new resource comprising satellite imagery, digital maps, and pixel-level labels of social semantic entities organized in a hierarchical structure. Additionally, we propose a novel vision-language reasoning framework called SocioReasoner that simulates the human process of identifying and annotating social semantic entities via cross-modal recognition and multi-stage reasoning. We employ reinforcement learning to optimize this non-differentiable process and elicit the reasoning capabilities of the vision-language model. Experiments demonstrate our approach's gains over state-of-the-art models and strong zero-shot generalization. The dataset and code are open-sourced under the Apache License 2.0 at https://github.com/AMAP-ML/SocioReasoner.

The rapid proliferation of misinformation across online platforms underscores the urgent need for robust, up-to-date, explainable, and multilingual fact-checking resources. However, existing datasets are limited in scope, often lacking multimodal evidence, structured annotations, and detailed links between claims, evidence, and verdicts. This paper introduces a comprehensive data collection and processing pipeline that constructs multimodal fact-checking datasets in French and German languages by aggregating ClaimReview feeds, scraping full debunking articles, normalizing heterogeneous claim verdicts, and enriching them with structured metadata and aligned visual content. We used state-of-the-art large language models (LLMs) and multimodal LLMs for (i) evidence extraction under predefined evidence categories and (ii) justification generation that links evidence to verdicts. Evaluation with G-Eval and human assessment demonstrates that our pipeline enables fine-grained comparison of fact-checking practices across different organizations or media markets, facilitates the development of more interpretable and evidence-grounded fact-checking models, and lays the groundwork for future research on multilingual, multimodal misinformation verification.

We introduce a novel FSVOS model that employs a local matching strategy to restrict the search space to the most relevant neighboring pixels. Rather than relying on inefficient standard im2col-like implementations (e.g., spatial convolutions, depthwise convolutions and feature-shifting mechanisms) or hardware-specific CUDA kernels (e.g., deformable and neighborhood attention), which often suffer from limited portability across non-CUDA devices, we reorganize the local sampling process through a direction-based sampling perspective. Specifically, we implement a non-parametric sampling mechanism that enables dynamically varying sampling regions. This approach provides the flexibility to adapt to diverse spatial structures without the computational costs of parametric layers and the need for model retraining. To further enhance feature coherence across frames, we design a supervised spatio-temporal contrastive learning scheme that enforces consistency in feature representations. In addition, we introduce a publicly available benchmark dataset for multi-object segmentation in X-ray angiography videos (MOSXAV), featuring detailed, manually labeled segmentation ground truth. Extensive experiments on the CADICA, XACV, and MOSXAV datasets show that our proposed FSVOS method outperforms current state-of-the-art video segmentation methods in terms of segmentation accuracy and generalization capability (i.e., seen and unseen categories). This work offers enhanced flexibility and potential for a wide range of clinical applications. Code is available at https://github.com/xilin-x/XRAVOS

Deformable scenes violate the rigidity assumptions underpinning classical visual--inertial odometry (VIO), often leading to over-fitting to local non-rigid motion or to severe camera pose drift when deformation dominates visual parallax. In this paper, we introduce DefVINS, the first visual-inertial odometry pipeline designed to operate in deformable environments. Our approach models the odometry state by decomposing it into a rigid, IMU-anchored component and a non-rigid scene warp represented by an embedded deformation graph. As a second contribution, we present VIMandala, the first benchmark containing real images and ground-truth camera poses for visual-inertial odometry in deformable scenes. In addition, we augment the synthetic Drunkard's benchmark with simulated inertial measurements to further evaluate our pipeline under controlled conditions. We also provide an observability analysis of the visual-inertial deformable odometry problem, characterizing how inertial measurements constrain camera motion and render otherwise unobservable modes identifiable in the presence of deformation. This analysis motivates the use of IMU anchoring and leads to a conditioning-based activation strategy that avoids ill-posed updates under poor excitation. Experimental results on both the synthetic Drunkard's and our real VIMandala benchmarks show that DefVINS outperforms rigid visual--inertial and non-rigid visual odometry baselines. Our source code and data will be released upon acceptance.

One of the main limitations of multirotor UAVs is their short flight time due to battery constraints. A practical solution for continuous operation is to power the drone from the ground via a tether. While this approach has been demonstrated for stationary systems, scenarios with a fast-moving base vehicle or strong wind conditions require modeling the tether forces, including aerodynamic effects. In this work, we propose two complementary approaches for real-time quasi-static tether modeling with aerodynamics. The first is an analytical method based on catenary theory with a uniform drag assumption, achieving very fast solve times below 1~ms. The second is a numerical method that discretizes the tether into segments and lumped masses, solving the equilibrium equations using CasADi and IPOPT. By leveraging initialization strategies, such as warm starting and analytical initialization, real-time performance was achieved with a solve time of 5~ms, while allowing for flexible force formulations. Both approaches were validated in real-world tests using a load cell to measure the tether force. The results show that the analytical method provides sufficient accuracy for most tethered UAV applications with minimal computational cost, while the numerical method offers higher flexibility and physical accuracy when required. These approaches form a lightweight and extensible framework for real-time tether simulation, applicable to both offline optimization and online tasks such as simulation, control, and trajectory planning.

Modern networks generate vast, heterogeneous traffic that must be continuously analyzed for security and performance. Traditional network traffic analysis systems, whether rule-based or machine learning-driven, often suffer from high false positives and lack interpretability, limiting analyst trust. In this paper, we present ReGAIN, a multi-stage framework that combines traffic summarization, retrieval-augmented generation (RAG), and Large Language Model (LLM) reasoning for transparent and accurate network traffic analysis. ReGAIN creates natural-language summaries from network traffic, embeds them into a multi-collection vector database, and utilizes a hierarchical retrieval pipeline to ground LLM responses with evidence citations. The pipeline features metadata-based filtering, MMR sampling, a two-stage cross-encoder reranking mechanism, and an abstention mechanism to reduce hallucinations and ensure grounded reasoning. Evaluated on ICMP ping flood and TCP SYN flood traces from the real-world traffic dataset, it demonstrates robust performance, achieving accuracy between 95.95% and 98.82% across different attack types and evaluation benchmarks. These results are validated against two complementary sources: dataset ground truth and human expert assessments. ReGAIN also outperforms rule-based, classical ML, and deep learning baselines while providing unique explainability through trustworthy, verifiable responses.

As Vision Transformers (ViTs) become standard vision backbones, a mechanistic account of their computational phenomenology is essential. Despite architectural cues that hint at dynamical structure, there is no settled framework that interprets Transformer depth as a well-characterized flow. In this work, we introduce the Block-Recurrent Hypothesis (BRH), arguing that trained ViTs admit a block-recurrent depth structure such that the computation of the original $L$ blocks can be accurately rewritten using only $k \ll L$ distinct blocks applied recurrently. Across diverse ViTs, between-layer representational similarity matrices suggest few contiguous phases. To determine whether these phases reflect genuinely reusable computation, we train block-recurrent surrogates of pretrained ViTs: Recurrent Approximations to Phase-structured TransfORmers (Raptor). In small-scale, we demonstrate that stochastic depth and training promote recurrent structure and subsequently correlate with our ability to accurately fit Raptor. We then provide an empirical existence proof for BRH by training a Raptor model to recover $96\%$ of DINOv2 ImageNet-1k linear probe accuracy in only 2 blocks at equivalent runtime. Finally, we leverage our hypothesis to develop a program of Dynamical Interpretability. We find i) directional convergence into class-dependent angular basins with self-correcting trajectories under small perturbations, ii) token-specific dynamics, where cls executes sharp late reorientations while patch tokens exhibit strong late-stage coherence toward their mean direction, and iii) a collapse to low rank updates in late depth, consistent with convergence to low-dimensional attractors. Altogether, we find a compact recurrent program emerges along ViT depth, pointing to a low-complexity normative solution that enables these models to be studied through principled dynamical systems analysis.

Synthetic Aperture Radar (SAR) imagery plays a critical role in all-weather, day-and-night remote sensing applications. However, existing SAR-oriented deep learning is constrained by data scarcity, while the physically grounded speckle noise in SAR imagery further hampers fine-grained semantic representation learning. To address these challenges, we propose SARMAE, a Noise-Aware Masked Autoencoder for self-supervised SAR representation learning. Specifically, we construct SAR-1M, the first million-scale SAR dataset, with additional paired optical images, to enable large-scale pre-training. Building upon this, we design Speckle-Aware Representation Enhancement (SARE), which injects SAR-specific speckle noise into masked autoencoders to facilitate noise-aware and robust representation learning. Furthermore, we introduce Semantic Anchor Representation Constraint (SARC), which leverages paired optical priors to align SAR features and ensure semantic consistency. Extensive experiments across multiple SAR datasets demonstrate that SARMAE achieves state-of-the-art performance on classification, detection, and segmentation tasks. Code and models will be available at https://github.com/MiliLab/SARMAE.

In this paper, we unleash the potential of the powerful monodepth model in camera-LiDAR calibration and propose CLAIM, a novel method of aligning data from the camera and LiDAR. Given the initial guess and pairs of images and LiDAR point clouds, CLAIM utilizes a coarse-to-fine searching method to find the optimal transformation minimizing a patched Pearson correlation-based structure loss and a mutual information-based texture loss. These two losses serve as good metrics for camera-LiDAR alignment results and require no complicated steps of data processing, feature extraction, or feature matching like most methods, rendering our method simple and adaptive to most scenes. We validate CLAIM on public KITTI, Waymo, and MIAS-LCEC datasets, and the experimental results demonstrate its superior performance compared with the state-of-the-art methods. The code is available at https://github.com/Tompson11/claim.

Modeling dexterous hand-object interactions is challenging as it requires understanding how subtle finger motions influence the environment through contact with objects. While recent world models address interaction modeling, they typically rely on coarse action spaces that fail to capture fine-grained dexterity. We, therefore, introduce DexWM, a Dexterous Interaction World Model that predicts future latent states of the environment conditioned on past states and dexterous actions. To overcome the scarcity of finely annotated dexterous datasets, DexWM represents actions using finger keypoints extracted from egocentric videos, enabling training on over 900 hours of human and non-dexterous robot data. Further, to accurately model dexterity, we find that predicting visual features alone is insufficient; therefore, we incorporate an auxiliary hand consistency loss that enforces accurate hand configurations. DexWM outperforms prior world models conditioned on text, navigation, or full-body actions in future-state prediction and demonstrates strong zero-shot transfer to unseen skills on a Franka Panda arm with an Allegro gripper, surpassing Diffusion Policy by over 50% on average across grasping, placing, and reaching tasks.

Multimodal Large Language Models have achieved impressive performance on a variety of vision-language tasks, yet their fine-grained visual perception and precise spatial reasoning remain limited. In this work, we introduce DiG (Differential Grounding), a novel proxy task framework where MLLMs learn fine-grained perception by identifying and localizing all differences between similar image pairs without prior knowledge of their number. To support scalable training, we develop an automated 3D rendering-based data generation pipeline that produces high-quality paired images with fully controllable discrepancies. To address the sparsity of difference signals, we further employ curriculum learning that progressively increases complexity from single to multiple differences, enabling stable optimization. Extensive experiments demonstrate that DiG significantly improves model performance across a variety of visual perception benchmarks and that the learned fine-grained perception skills transfer effectively to standard downstream tasks, including RefCOCO, RefCOCO+, RefCOCOg, and general multimodal perception benchmarks. Our results highlight differential grounding as a scalable and robust approach for advancing fine-grained visual reasoning in MLLMs.

Feed-forward 3D reconstruction models such as DUSt3R, VGGT, and Depth Anything 3 (DA3) are transformer-based foundation models that infer camera geometry and dense scene structure in a single forward pass. Trained at scale in a supervised fashion, they raise a central question: do these models build upon geometric principles akin to traditional multi-view pipelines, or do they primarily rely on learned priors arising from the large-scale training setup? We find that epipolar geometry emerges within the intermediate layers of all three models and is causally linked to correspondence patterns in attention heads. To study this, we perform a systematic analysis of their internal representations across three real-world datasets and a controlled synthetic dataset. We quantify geometric understanding by probing intermediate features, analyzing attention patterns to identify correspondence matching patterns, and performing targeted interventions at the attention level. Further, we assess the role of learned priors by applying challenging input-level perturbations, such as occlusions, scene ambiguities, and varying camera configurations, and compare them against classical multi-stage reconstruction pipelines.

Existing methods achieve high-quality facial albedo capture under controllable lighting, which increases capture cost and limits usability. We propose WildCap, a novel method for high-quality facial albedo capture from a smartphone video recorded in the wild. To disentangle high-quality albedo from complex lighting effects in in-the-wild captures, we propose a novel hybrid inverse rendering framework. We first apply a data-driven method, i.e., SwitchLight, to convert the captured images into more constrained conditions and then adopt model-based inverse rendering. However, unavoidable local artifacts in network predictions, such as shadow-baking, are non-physical and thus hinder accurate inverse rendering of lighting and material. To address this, we propose a novel texel grid lighting model to explain non-physical effects as clean albedo illuminated by local physical lighting. During optimization, we jointly sample a diffusion prior for the albedo map and optimize the lighting, effectively resolving scale ambiguity between local lights and albedo. Other reflectance maps are then predicted from the albedo. Our method achieves significantly better results than prior arts in the same capture setup, closing the quality gap between in-the-wild and controllable recordings by a large margin.

Training vision models with language supervision enables general and transferable representations. However, many visual domains, especially non-object-centric domains such as medical imaging and remote sensing, contain itemized text annotations: multiple text items describing distinct and semantically independent findings within a single image. Such supervision differs from standard multi-caption supervision, where captions are redundant or highly overlapping. Here, we introduce ItemizedCLIP, a framework for learning complete and explainable visual representations from itemized text supervision. ItemizedCLIP employs a cross-attention module to produce text item-conditioned visual embeddings and a set of tailored objectives that jointly enforce item independence (distinct regions for distinct items) and representation completeness (coverage of all items). Across four domains with naturally itemized text supervision (brain MRI, head CT, chest CT, remote sensing) and one additional synthetically itemized dataset, ItemizedCLIP achieves substantial improvements in zero-shot performance and fine-grained interpretability over baselines. The resulting ItemizedCLIP representations are semantically grounded, item-differentiable, complete, and visually interpretable. Our code is available at https://github.com/MLNeurosurg/ItemizedCLIP.

Stereo foundation models achieve strong zero-shot generalization but remain computationally prohibitive for real-time applications. Efficient stereo architectures, on the other hand, sacrifice robustness for speed and require costly per-domain fine-tuning. To bridge this gap, we present Fast-FoundationStereo, a family of architectures that achieve, for the first time, strong zero-shot generalization at real-time frame rate. We employ a divide-and-conquer acceleration strategy with three components: (1) knowledge distillation to compress the hybrid backbone into a single efficient student; (2) blockwise neural architecture search for automatically discovering optimal cost filtering designs under latency budgets, reducing search complexity exponentially; and (3) structured pruning for eliminating redundancy in the iterative refinement module. Furthermore, we introduce an automatic pseudo-labeling pipeline used to curate 1.4M in-the-wild stereo pairs to supplement synthetic training data and facilitate knowledge distillation. The resulting model can run over 10x faster than FoundationStereo while closely matching its zero-shot accuracy, thus establishing a new state-of-the-art among real-time methods. Project page: https://nvlabs.github.io/Fast-FoundationStereo/

Many manufacturing environments operate in low-light conditions or within enclosed machines where conventional vision systems struggle. Infrared cameras provide complementary advantages in such environments. Simultaneously, supervised AI systems require large labeled datasets, which makes zero-shot learning frameworks more practical for applications including infrared cameras. Recent advances in vision-language foundation models (VLMs) offer a new path in zero-shot predictions from paired image-text representations. However, current VLMs cannot understand infrared camera data since they are trained on RGB data. This work introduces VLM-IRIS (Vision-Language Models for InfraRed Industrial Sensing), a zero-shot framework that adapts VLMs to infrared data by preprocessing infrared images captured by a FLIR Boson sensor into RGB-compatible inputs suitable for CLIP-based encoders. We demonstrate zero-shot workpiece presence detection on a 3D printer bed where temperature differences between the build plate and workpieces make the task well-suited for thermal imaging. VLM-IRIS converts the infrared images to magma representation and applies centroid prompt ensembling with a CLIP ViT-B/32 encoder to achieve high accuracy on infrared images without any model retraining. These findings demonstrate that the proposed improvements to VLMs can be effectively extended to thermal applications for label-free monitoring.

Multi-Agent Path Finding (MAPF) algorithms are increasingly deployed in industrial warehouses and automated manufacturing facilities, where robots must operate reliably under real-world physical constraints. However, existing MAPF evaluation frameworks typically rely on simplified robot models, leaving a substantial gap between algorithmic benchmarks and practical performance. Recent frameworks such as SMART combine kinodynamic modeling with execution based on the Action Dependency Graph (ADG), enabling realistic, large-scale MAPF evaluation. Building on this capability, this work investigates how key planner design choices influence performance under realistic execution settings. We systematically study three fundamental factors: (1) the relationship between solution optimality and execution performance, (2) the sensitivity of system performance to inaccuracies in kinodynamic modeling, and (3) the tradeoff between model accuracy and plan optimality. Empirically, we examine these factors to understand how these design choices affect performance in realistic scenarios. We highlight open challenges and research directions to steer the community toward practical, real-world deployment.

We present COREA, the first unified three-tasks framework that couples an SDF and relightable 3D Gaussians (3DGS) to jointly support SH-based novel-view synthesis (NVS), surface reconstruction, and inverse physically-based rendering (inverse PBR). While recent relightable 3DGS methods have progressed, inverse PBR remains bottlenecked by normal estimation, as the discrete nature of 3DGS often yields oversmoothed and unstable normals. To address this limitation, COREA couples the complementary geometric properties of an SDF and relightable 3DGS on a shared underlying surface, where geometry-constrained relightable 3DGS provides reliable depth signals to anchor SDF geometry and the continuous SDF normal field provides spatially consistent supervision for Gaussian normal learning. We couple these signals through depth-guided alignment and normal supervision with normal-aware densification, and introduce Dual-Density Control to regulate densification by balancing photometric and geometric gradients for stable, memory-efficient training. Experiments on standard benchmarks show that COREA is the only framework that supports all three tasks, achieving competitive performance overall, with particularly superior results in inverse PBR.

Pansharpening fuses a high-resolution panchromatic (PAN) image with a low-resolution multispectral (LRMS) image to produce a high-resolution multispectral (HRMS) image. A key difficulty is that jointly processing PAN and MS features often entangles spatial detail enhancement with spectral fidelity. To address this feature entanglement, we propose S2WMamba, a framework that explicitly disentangles modality-specific frequency information for highly controlled crossmodal interaction. Concretely, unlike global frequency transforms, a localized 2D Haar DWT is applied to the PAN image to precisely isolate spatial edges and textures. Concurrently, a novel channel-wise 1D Haar DWT treats each pixel's spectrum as a 1D signal, isolating the shared spectral base from band-specific variations to strictly limit spectral distortion. The resulting Spectral branch injects wavelet-extracted spatial details into MS features, while the Spatial branch refines PAN features using spectra from the DWT1D process. To overcome inadequate frequency fusion, the two branches exchange information via Mambabased cross-modulation, which explicitly models long-range dependencies across these decoupled sub-bands with linear complexity. On WV3, GF2, and QB datasets, S2WMamba matches or surpasses recent strong baselines (FusionMamba, CANNet, U2Net, PanNet), improving PSNR by up to 0.23 dB and reaching an HQNR of 0.956 on full-resolution WV3. Extensive ablations justify the modality-specific DWT placement and the parallel dual-branch architecture.

Real-time monocular 3D object detection remains challenging due to severe depth ambiguity, viewpoint shifts, and the high computational cost of 3D reasoning. Existing approaches either rely on LiDAR or geometric priors to compensate for missing depth or sacrifice efficiency to achieve competitive accuracy. We introduce LeAD-M3D, a monocular 3D detector that achieves state-of-the-art accuracy and real-time inference without extra modalities. Our method is enabled by three key components. Asymmetric Augmentation Denoising Distillation (A2D2) transfers geometric knowledge from a clean-image teacher to a MixUp-noised student via a quality- and importance-weighted depth-feature loss, enabling stronger depth reasoning without LiDAR. 3D-aware Consistent Matching (CM$_{\text{3D}}$) improves prediction-to-ground truth assignment by integrating 3D MGIoU into the matching score, yielding stable and precise supervision. Finally, Confidence-Gated 3D Inference (CGI$_{\text{3D}}$) accelerates inference by restricting expensive 3D regression to confident regions. Together, these contributions set a new Pareto frontier for monocular 3D detection: LeAD-M3D achieves state-of-the-art accuracy on KITTI and Waymo, and the best reported car AP on Rope3D, while running up to 3.6$\,\times$ faster than prior high-accuracy models (e.g., MonoDiff). LeAD-M3D demonstrates that high fidelity and real-time monocular 3D detection is simultaneously attainable, without LiDAR, stereo, or strong geometric assumptions.

VR sketching lets users explore and iterate on ideas directly in 3D, offering a faster and more intuitive alternative to conventional CAD tools. However, existing sketch-to-shape models ignore the temporal ordering of strokes, discarding crucial cues about structure and design intent. We introduce VRSketch2Shape, the first framework and multi-category dataset for generating 3D shapes from sequential VR sketches. Our contributions are threefold: (i) an automated pipeline that generates sequential VR sketches from arbitrary shapes, (ii) a dataset of over 20k synthetic and 900 hand-drawn sketch-shape pairs across four categories, and (iii) an order-aware sketch encoder coupled with a diffusion-based 3D generator. Our approach yields higher geometric fidelity than prior work, generalizes effectively from synthetic to real sketches with minimal supervision, and performs well even on partial sketches. All data and models will be released open-source at https://chenyizi086.github.io/VRSketch2Shape_website.

The role of phase in neural sequence models remains poorly understood. To isolate this question, we introduce PRISM, a complex-valued encoder that enforces a unit-norm constraint ($|z| = 1$) and replaces attention with gated spectral filtering. Under this constraint, the model cannot use activation magnitude to distinguish signal from noise, and must instead rely on phase angles. We find that semantic relationships correlate with measurable phase structure: synonym pairs exhibit significantly higher phase coherence than random pairs ($R = 0.198$ vs.\ $0.072$, $p < 0.001$), and the model resolves lexical ambiguity via layer-specific phase rotations while maintaining near-unit gain. These phase representations are robust to scalar attenuation, retaining $97\%$ of translation quality when signal magnitude is uniformly reduced. We also identify a spectral density threshold: the model fails to generate coherent output from isolated tokens, requiring minimum sequence length to produce the interference patterns that support its computation. Finally, we show that a hybrid architecture (Wave-Particle Transformer) combining a phase-based encoder with standard attention matches Transformer baselines at $33$M parameters with fewer non-embedding parameters, though we do not claim this generalizes to larger scales. Our findings provide controlled evidence that phase angles can encode semantic information in complex-valued networks, and characterize the conditions under which this encoding succeeds and fails.

Synthetic data generation is an important tool for privacy-preserving data sharing. Although diffusion models have set recent benchmarks, flow matching (FM) offers a promising alternative. This paper presents different ways to implement FM for tabular data synthesis. We provide a comprehensive empirical study that compares flow matching (FM and variational FM) with a state-of-the-art diffusion method (TabDDPM and TabSyn) in tabular data synthesis. We evaluate both the standard Optimal Transport (OT) and the Variance Preserving (VP) probability paths, and also compare deterministic and stochastic samplers -- something possible when learning to generate using \textit{variational} FM -- characterising the empirical relationship between data utility and privacy risk. Our key findings reveal that FM, particularly TabbyFlow, outperforms diffusion baselines. Flow matching methods also achieve better performance with remarkably low function evaluations ($\leq$ 100 steps), offering a substantial computational advantage. The choice of probability path is also crucial, as using the OT is a strong default and more robust to early stopping on average, while VP has potential to produce synthetic data with lower privacy risk. Lastly, our results show that making flows stochastic not only preserves marginal distributions but, in some instances, enables the generation of high utility synthetic data with reduced disclosure risk. The implementation code associated with this paper is publicly available at https://github.com/rulnasution/tabular-flow-matching.

Visual-Inertial Odometry (VIO) is a critical component for robust ego-motion estimation, enabling foundational capabilities such as autonomous navigation in robotics and real-time 6-DoF tracking for augmented reality. Existing methods face a well-known trade-off: filter-based approaches are efficient but prone to drift, while optimization-based methods, though accurate, rely on computationally prohibitive Visual-Inertial Bundle Adjustment (VIBA) that is difficult to run on resource-constrained platforms. Rather than removing VIBA altogether, we aim to reduce how often and how heavily it must be invoked. To this end, we cast two key design choices in modern VIO, when to run the visual frontend and how strongly to trust its output, as sequential decision problems, and solve them with lightweight reinforcement learning (RL) agents. Our framework introduces a lightweight, dual-pronged RL policy that serves as our core contribution: (1) a Select Agent intelligently gates the entire VO pipeline based only on high-frequency IMU data; and (2) a composite Fusion Agent that first estimates a robust velocity state via a supervised network, before an RL policy adaptively fuses the full (p, v, q) state. Experiments on the EuRoC MAV and TUM-VI datasets show that, in our unified evaluation, the proposed method achieves a more favorable accuracy-efficiency-memory trade-off than prior GPU-based VO/VIO systems: it attains the best average ATE while running up to 1.77 times faster and using less GPU memory. Compared to classical optimization-based VIO systems, our approach maintains competitive trajectory accuracy while substantially reducing computational load.

Group Relative Policy Optimization (GRPO) enables stable and preference-oriented updates via group-wise comparisons for post-training video generation. However, GRPO directly optimizes reward-induced advantages. Under sustained optimization, the reward score can lose fidelity as a proxy for true video quality, consistent with the phenomenon described by Goodhart's Law. This leads to two recurring issues: (i) shortcut-driven optimization under composite objectives and (ii) reward saturation within prompt groups. To address these issues, we introduce TaRoS, a Target-Robust Reward Signaling framework for Video generation GRPO. TaRoS leverages component level performance assessment together with intra-group sparsity to organize multi-aspect rewards towards optimization objectives. In addition, it adaptively downweights components that exhibit saturation, thereby preserving effective optimization directions and mitigating redundancy. This maintains meaningful optimization directions and preserves within-group ranking separation, thereby preventing reward hacking and leading to more reliable policy updates. Extensive experiments show consistent improvements in visual fidelity, motion coherence, and text-video alignment over strong baselines.