Combining the Color and Event cameras (also called Dynamic Vision Sensors, DVS) for robust object tracking is a newly emerging research topic in recent years. Existing color-event tracking framework usually contains multiple scattered modules which may lead to low efficiency and high computational complexity, including feature extraction, fusion, matching, interactive learning, etc. In this paper, we propose a single-stage backbone network for Color-Event Unified Tracking (CEUTrack), which achieves the above functions simultaneously. Given the event points and RGB frames, we first transform the points into voxels and crop the template and search regions for both modalities, respectively. Then, these regions are projected into tokens and parallelly fed into the unified Transformer backbone network. The output features will be fed into a tracking head for target object localization. Our proposed CEUTrack is simple, effective, and efficient, which achieves over 75 FPS and new SOTA performance. To better validate the effectiveness of our model and address the data deficiency of this task, we also propose a generic and large-scale benchmark dataset for color-event tracking, termed COESOT, which contains 90 categories and 1354 video sequences. Additionally, a new evaluation metric named BOC is proposed in our evaluation toolkit to evaluate the prominence with respect to the baseline methods. We hope the newly proposed method, dataset, and evaluation metric provide a better platform for color-event-based tracking. The dataset, toolkit, and source code will be released on: \url{https://github.com/Event-AHU/COESOT}.

The traditional paradigm to update retrieval models requires re-computing the embeddings of the gallery data, a time-consuming and computationally intensive process known as backfilling. To circumvent backfilling, Backward-Compatible Learning (BCL) has been widely explored, which aims to train a new model compatible with the old one. Many previous works focus on effectively aligning the embeddings of the new model with those of the old one to enhance the backward-compatibility. Nevertheless, such strong alignment constraints would compromise the discriminative ability of the new model, particularly when different classes are closely clustered and hard to distinguish in the old feature space. To address this issue, we propose to relax the constraints by introducing perturbations to the old feature prototypes. This allows us to align the new feature space with a pseudo-old feature space defined by these perturbed prototypes, thereby preserving the discriminative ability of the new model in backward-compatible learning. We have developed two approaches for calculating the perturbations: Neighbor-Driven Prototype Perturbation (NDPP) and Optimization-Driven Prototype Perturbation (ODPP). Particularly, they take into account the feature distributions of not only the old but also the new models to obtain proper perturbations along with new model updating. Extensive experiments on the landmark and commodity datasets demonstrate that our approaches perform favorably against state-of-the-art BCL algorithms.

Learned image compression (LIC) has shown great promise for achieving high rate-distortion performance. However, current LIC methods are often limited in their capability to model the complex correlation structures inherent in natural images, particularly the entanglement of invariant global structures with transient local textures within a single monolithic representation. This limitation precipitates severe geometric deformation at low bitrates. To address this, we introduce a framework predicated on functional decomposition, which we term Deeply-Conditioned Image Compression via self-generated priors (DCIC-sgp). Our central idea is to first encode a potent, self-generated prior to encapsulate the image’s structural backbone. This prior is subsequently utilized not as mere side-information, but to holistically modulate the entire compression pipeline. This deep conditioning, most critically of the analysis transform, liberates it to dedicate its representational capacity to the residual, high-entropy details. This hierarchical, dependency-driven approach achieves an effective disentanglement of information streams. Our extensive experiments validate this assertion; visual analysis demonstrates that our method substantially mitigates the geometric deformation artifacts that plague conventional codecs at low bitrates. Quantitatively, our framework establishes highly competitive performance, achieving significant BD-rate reductions of 14.4 %, 15.7 %, and 15.1 % against the VVC test model VTM-12.1 on the Kodak, CLIC, and Tecnick datasets.

In this study, we introduce EinsPT, an efficient instance-aware pre-training paradigm designed to reduce the transfer gap between vision foundation models and downstream instance-level tasks. Unlike conventional image-level pre-training that relies solely on unlabeled images, EinsPT leverages both image reconstruction and instance annotations to learn representations that are spatially coherent and instance discriminative. To achieve this efficiently, we propose a proxy-foundation architecture that decouples high-resolution and low-resolution learning: the foundation model processes masked low-resolution images for global semantics, while a lightweight proxy model operates on complete high-resolution images to preserve fine-grained details. The two branches are jointly optimized through reconstruction and instance-level prediction losses on fused features. Extensive experiments demonstrate that EinsPT consistently enhances recognition accuracy across various downstream tasks with substantially reduced computational cost, while qualitative results further reveal improved instance perception and completeness in visual representations. Code is available at github.com/feufhd/EinsPT.

Although Multimodal Large Language Models (MLLMs) have shown remarkable generalization across diverse vision-language tasks, recent studies reveal their limitations in visual discrimination. These challenges arise not from insufficient model capacity, but from existing training paradigms that favor linguistic priors over detailed visual analysis. While existing approaches address this limitation through external interventions such as feature integration or knowledge augmentation, we propose a Group-Relative Visual Discrimination Enhancement framework to unlock intrinsic capability of MLLMs and requires no external resources. Our method introduces a Group-Relative Reinforcement Learning paradigm equipped with a lightweight Visual Patch Selection Plugin to dynamically select discriminative visual tokens. The framework establishes a self-feedback loop between visual encoder and language decoder, leveraging the dual reward-penalty signals derived from the model’s internal language feedback to optimize the visual focus, thereby enhancing the model’s visual discrimination capabilities. Extensive experimental results across six visual recognition benchmarks and two VQA benchmarks demonstrate the effectiveness of our method. Code is available at https://github.com/FannierPeng/GROVE.

Object tracking is a fundamental task in computer vision, and ensuring its security is crucial for reliable and safe real-world applications. Similar background distractors are commonly present in tracking scenarios and often result in tracking failures. However, tracking attack methods leveraging background similarity have yet to receive adequate attention. In this paper, we propose a saliency-aware guided adversarial attack method for tracking models. The proposed method aims to evaluate the performance of advanced trackers under adversarial attacks, targeting interference from similar objects to reveal their vulnerabilities. Specifically, we introduce a text-guided similar region localization module that leverages semantic features of coarse-grained category attributes to select regions from multiple frames that are both visually and semantically similar to the target, overcoming the limitations of random selection methods. In addition, we develop a saliency-aware search region feature constraint module that incorporates foreground-background saliency information. By dynamically filtering feature weights through the joint use of fine-grained target appearance and text descriptions, this adaptive mechanism reduces the confidence of the tracker in foreground-background distinctions, progressively leading to target drift. We conduct extensive experiments on multiple benchmark datasets over multiple types of trackers, and the results demonstrate that the proposed method can effectively attack existing trackers and reveal their potential security risks.

Linear attention significantly reduces the computational complexity of Transformers from quadratic to linear, yet it consistently lags behind softmax-based attention in performance. We identify the root cause of this degradation as the non-negativity constraint imposed on kernel feature maps: standard projections like ReLU act as "passive truncation" operators, indiscriminately discarding semantic information residing in the negative domain. We propose MirrorLA, a geometric framework that substitutes passive truncation with active reorientation. By leveraging learnable Householder reflections, MirrorLA rotates the feature geometry into the non-negative orthant to maximize information retention. Our approach restores representational density through a cohesive, multi-scale design: it first optimizes local discriminability via block-wise isometries, stabilizes long-context dynamics using variance-aware modulation to diversify activations, and finally, integrates dispersed subspaces via cross-head reflections to induce global covariance mixing. MirrorLA achieves state-of-the-art performance across standard benchmarks, demonstrating that strictly linear efficiency can be achieved without compromising representational fidelity.

In text based video retrieval, text queries commonly convey certain objects and events the user desires to retrieve, which are more concise cues than video contents. Distinct information density between video contents and query cues leads to the difficulty of query-video alignment. To pursue a more compact video representation and accurate textual-visual feature matching, this paper introduces a novel VideoAligner to disentangle video features. VideoAligner first generates ‘object’ and ‘event’ tokens from query texts. It subsequently spots and merges visual tokens related to concepts in the query. In other words, we use ‘object’ and ‘event’ tokens to represent cues of query, which therefore supervise the disentanglement and extraction of meaningful visual features from videos. VideoAligner finally leads to compact visual tokens explicitly depicting query objects and events. Extensive experiments on three widely-used datasets demonstrate the promising performance and domain generalization capability of our method. For instance, our method shows better efficiency and consistently outperforms many recent works like ProST on three datasets. We hope to inspire future work for collaborative cross-modal learning with certain modality as guidance.

Existing visual trackers mainly operate in a non-interactive, fire-and-forget manner, making them impractical for real-world scenarios that require human-in-the-loop adaptation. To overcome this limitation, we introduce Interactive Tracking, a new paradigm that allows users to guide the tracker at any time using natural language commands. To support research in this direction, we make three main contributions. First, we present InteractTrack, the first large-scale benchmark for interactive tracking, containing 150 videos with dense bounding box annotations and timestamped language instructions. Second, we propose a comprehensive evaluation protocol and evaluate 25 representative trackers, showing that state-of-the-art methods fail in interactive scenarios—strong performance on conventional benchmarks does not transfer. Third, we introduce Interactive Memory-Augmented Tracking (IMAT), a new baseline that employs a dynamic memory mechanism to learn from user feedback and update tracking behavior accordingly. Our benchmark, protocol, and baseline establish a foundation for developing more intelligent, adaptive, and collaborative tracking systems, bridging the gap between automated perception and human guidance.

While 3D Gaussian Splatting (3DGS) has demonstrated impressive real-time rendering performance, its efficacy remains constrained by a reliance on heuristic density control. Despite numerous refinements to these handcrafted rules, such methods inherently lack the flexibility to adapt to diverse scenes with complex geometries. In this paper, we propose a paradigm shift for density control from rigid heuristics to fully learnable policies. Specifically, we introduce LeGS, a framework that reformulates density control as a parameterized policy network optimized via Reinforcement Learning (RL). Central to our approach is the tailored effective reward function grounded in sensitivity analysis, which precisely quantifies the marginal contribution of individual Gaussians to reconstruction quality. To maintain computational tractability, we derive a closed-form solution that reduces the complexity of reward calculation from O(N^2) to O(N). Extensive experiments on the Mip-NeRF 360, Tanks & Temples, and Deep Blending datasets demonstrate that LeGS significantly outperforms state-of-the-art methods, striking a superior balance between reconstruction quality and efficiency. The code will be released at https://github.com/AaronNZH/LeGS

While large language models provide strong compositional reasoning, existing reasoning segmentation pipelines fail to transparently connect this reasoning to visual perception.Current methods, such as latent query alignment, are end-to-end yet opaque "black boxes".Conversely, textual localization readout is merely readable, not truly interpretable, often functioning as an unconstrained post-hoc step.To bridge this interpretability gap, we propose SegCompass, an end-to-end model that leverages a Sparse Autoencoder (SAE) to forge an explicit, interpretable, and differentiable alignment pathway.Given an image-instruction pair, SegCompass first generates a chain-of-thought (CoT) trace.The core of our method is an SAE that maps both the CoT and visual tokens into a shared, high-dimensional sparse concept space.A query codebook selects salient concepts from this space, which are then spatially grounded by a slot mapper into a multi-slot heatmap that guides the final mask decoder.The entire model is trained jointly, unifying reinforcement learning for the reasoning path with standard segmentation supervision.This SAE-driven interface provides a "white-box" connection that is significantly more traceable than latent queries and more coherent than textual readouts.Extensive experiments on five challenging benchmarks demonstrate that SegCompass matches or surpasses state-of-the-art performance.Crucially, our visual and quantitative analyses show a strong correlation between the quality of the learned sparse concepts and final mask accuracy, confirming that SegCompass achieves superior results through its enhanced and inspectable alignment.Faithful code will be released publicly.

Instruction finetuning of Multimodal Large Language Models (MLLMs) relies on massive datasets, hindering scalability and efficiency. Consequently, data pruning has emerged as a vital strategy to attain full-dataset performance using a minimal subset. However, existing methods often incur high computational overheads or suffer from performance collapse at high pruning rates. To address these challenges, we propose Decoupled Gradient-Guided Stratification (DGGS), a resource-efficient data pruning framework. Diverging from approaches that introduce additional computation via external proxy models, we natively repurpose the internal vision encoder to stratify data based on visual complexity, and employ the LLM module as a deterministic verifier to quantify the semantic density of each stratum. DGGS dynamically allocates the pruning budget to stratified regions where semantic importance outweighs visual complexity, filtering out visual noise. Empirically, DGGS operates with minimal additional memory overhead and achieves robust performance, retaining approximately 85% of full-dataset performance with 1% of the training data.

Cycle self-training (CST) breaks the shared classifier assumption of the standard self-training framework, which is effective for unsupervised domain adaptation and exploits unlabeled target data by training with target pseudo-labels. CST introduces a target classifier and employs an inner-outer loop updating strategy, addressing the issue of unreliable pseudo-labels and enabling pseudo-labels to generalize across domains. Despite its success in image classification, extending CST to object detection faces three main challenges. First, the upper bound of CST in object detection is constrained by three types of unreliable pseudo-labels, such as classification error alone, localization error alone, and their combination. Second, since object detection involves detecting multiple target objects, directly applying CST leads to training insta bility. Third, a wider numerical range of regression coordinates leads to exploding losses. To this end, we apply CST to both classification and regression and propose the Dual-Stream Bilevel-Cycle Optimization framework. Specifically, we construct CST upon Mean Teacher to prevent training instability and use extra normalization to map the regression bounding box into a standardized space, effectively addressing exploding losses. Also, we provide a theoretical derivation of the regression bound. Extensive experiments across four cross domain standard scenarios demonstrate that our framework achieves considerable results.

With the emergence of wireless applications in three-dimensional environments, such as the low-altitude airspace and 3D heterogeneous networks, radio map estimation is increasingly required to characterize signal propagation across both horizontal and vertical dimensions. However, extending radio map estimation from 2D to 3D remains challenging due to increased spatial sparsity and limited supervision across continuous altitudes. In this paper, we propose RadioFormer3D, a specialized model for volumetric spectrum reconstruction under weak supervision. Building on the dual-stream, multi-granularity fusion architecture of RadioFormer, RadioFormer3D introduces a Fourier-based sampling encoder and a volumetric decoder to efficiently process sparse measurements in 3D space. To alleviate the lack of vertical supervision, we propose the Joint Spectrum Integrity Loss, which integrates volume-level pseudo-label supervision, map-level geometry-aware radio rendering, and pixel-level localized constraints within a unified optimization scheme. This design enables the model to capture complex vertical structural relationships more effectively under sparse supervision. Extensive experiments across several radio map datasets show that RadioFormer3D achieves superior overall performance compared to representative existing methods. In particular, it demonstrates improved reconstruction quality at unlabeled altitudes while maintaining a favorable trade-off between accuracy and inference efficiency, positioning it as a highly promising solution for future 3D environment-aware wireless networks.

Vision-language tracking guided by natural language specifications leverages high-level semantic cues of target objects to substantially boost tracking accuracy and robustness. Existing studies have verified that adaptively optimizing textual descriptions throughout the tracking process can effectively mitigate the semantic-visual mismatch induced by dynamic variations in target appearance, position, and other inherent attributes. Nevertheless, mainstream methods that directly generate textual information via sequence models or large language models inevitably suffer from inherent defects, including erroneous target updating, excessive background distraction, and pervasive hallucination artifacts. To address the aforementioned limitations, this paper proposes a novel language dependency parsing mechanism to precisely distill core tracking principal components, encompassing target objects, semantic concepts, and background contextual information. On this basis, we perform component-aware adaptive textual description updates by exploiting the powerful cross-modal understanding capability of the pre-trained vision-language model Qwen-VL. By integrating the proposed elaborately designed modules into the baseline framework, our method achieves consistent and superior tracking performance on multiple large-scale vision-language tracking benchmarks, including TNL2K, LaSOT, TNLLT, and OTB-LANG. The source code and pre-trained models will be released at https://github.com/Event-AHU/Open_VLTrack.

Vision-Language Navigation in Continuous Environments (VLN-CE) requires agents to follow natural-language instructions while navigating in real-world-like environments. Most VLN-CE approach­es adopt a three-stage framework: a waypoint predictor proposes navigable waypoints, and a navigator selects the best waypoint, with a low-level controller executing the movement to it. However, this decoupled paradigm often leads to unreachable waypoints or inconsistencies between planning and control. In this work, instead of predicting isolated waypoints, we introduce a novel paradigm called Trajectory Waypoint, which grounds each candidate waypoint in an executable trajectory. To realize this, we design a Trajectory Waypoint Predictor formulated as a TSDF-guided diffusion policy, which steers trajectory generation away from obstacles, inherently ensuring the reachability of the predicted waypoints. We further propose a trajectory-enhanced navigator that injects the associated trajectory as additional information for planning, enabling strict consistency between high-level semantic decisions and low-level execution. Extensive experiments on the VLN-CE benchmark show that our Trajectory Waypoint paradigm achieves superior performance over the baselines.

Deep neural networks have achieved success, but require enormous computational costs and storage space on large-scale datasets. Dataset pruning is an effective method to construct compact subsets from large-scale datasets to reduce computational costs while preserving model performance. Existing dataset pruning methods predominantly focus on clean benchmark datasets, overlooking the critical challenges of real-world scenarios with prevalent noisy label samples. In this paper, we find that previous methods have the limitation of being unable to distinguish noisy label samples from hard samples, resulting in overfitting to noisy label samples. To address this, we propose Robust Pruning (RoP), a two-stage dataset pruning framework that consists of noisy label discrimination and re-labeling. Specifically, RoP employs the divergence in label distributions between given labeled samples and their neighboring predictions as an importance metric to identify noisy samples. Compared to other neighborhood-based methods, RoP corrects uncertain neighboring samples through feature and label propagation. Subsequently, RoP applies re-labeling techniques to the selected subset to prevent overfitting of noisy labels. Finally, RoP evaluates easy and hard samples through empirical analysis of re-labeling accuracy, guaranteeing the coverage of hard and easy samples in the selected subset. Extensive experiments demonstrate the effectiveness of the proposed method, not only on real-world benchmarks but also on synthetic datasets, highlighting its suitability for noisy label scenarios. The code is available at https://github.com/successhaha/RoP.

X-ray image based medical report generation achieves significant progress in recent years with the help of large language models, however, these models have not fully exploited the effective information in visual image regions, resulting in reports that are linguistically sound but insufficient in describing key diseases. In this paper, we propose a novel associative memory-enhanced X-ray report generation model that effectively mimics the process of professional doctors writing medical reports. It considers both the mining of global and local visual information and associates historical report information to better complete the writing of the current report. Specifically, given an X-ray image, we first utilize a classification model along with its activation maps to accomplish the mining of visual regions highly associated with diseases and the learning of disease query tokens. Then, we employ a visual Hopfield network to establish memory associations for disease-related tokens, and a report Hopfield network to retrieve report memory information. This process facilitates the generation of high-quality reports based on a large language model and achieves state-of-the-art performance on multiple benchmark datasets, including the IU X-ray, MIMIC-CXR, and Chexpert Plus. The source code and pre-trained models of this work have been released on https://github.com/Event-AHU/Medical_Image_Analysis.

Unlike conventional visual question answering, video-grounded dialog requires a deep understanding of both the dialog history and the video content to generate accurate responses. Although existing methods have achieved promising results, they still struggle with progressively comprehending complex dialog history and effectively integrating video information. To address these challenges, we propose an iterative search and reasoning framework composed of a textual encoder, a visual encoder, and a generator. Specifically, the textual encoder adopts a path search and aggregation strategy to identify key cues in the dialog history that are essential for understanding the current question. Meanwhile, the visual encoder employs an iterative reasoning network to extract and highlight critical visual evidence from the video, thereby enabling more comprehensive visual understanding. Finally, we use a pretrained GPT-2 model as the answer generator to transform the discovered latent cues into coherent and contextually appropriate responses. Extensive experiments on three public datasets demonstrate the effectiveness and generalizability of the proposed framework.