论文检索

Xuchen Li, Xuzhao Li, Shiyu Hu, Kaiqi Huang, Wentao Zhang

Recent advances in large language models (LLMs) have improved reasoning in text and image domains, yet achieving robust video reasoning remains a significant challenge. Existing video benchmarks mainly assess shallow understanding and reasoning and allow models to exploit global context, failing to rigorously evaluate true causal and stepwise reasoning. We present CausalStep, a benchmark designed for explicit stepwise causal reasoning in videos. CausalStep segments videos into causally linked units and enforces a strict stepwise question-answer (QA) protocol, requiring sequential answers and preventing shortcut solutions. Each question includes carefully constructed distractors based on error type taxonomy to ensure diagnostic value. The benchmark features 100 videos across six categories and 1,852 multiple-choice QA pairs. We introduce seven diagnostic metrics for comprehensive evaluation, enabling precise diagnosis of causal reasoning capabilities. Experiments with leading proprietary and open-source models, as well as human baselines, reveal a significant gap between current models and human-level stepwise reasoning. CausalStep provides a rigorous benchmark to drive progress in robust and interpretable video reasoning.

Xilai Li, Xiaosong Li, Weijun Jiang

Multi-modality image fusion enhances scene perception by combining complementary information. Unified models aim to share parameters across modalities for multi-modality image fusion, but large modality differences often cause gradient conflicts, limiting performance. Some methods introduce modality-specific encoders to enhance feature perception and improve fusion quality. However, this strategy reduces generalisation across different fusion tasks. To overcome this limitation, we propose a unified multi-modality image fusion framework based on channel perturbation and pre-trained knowledge integration (UP-Fusion). To suppress redundant modal information and emphasize key features, we propose the Semantic-Aware Channel Pruning Module (SCPM), which leverages the semantic perception capability of a pre-trained model to filter and enhance multi-modality feature channels. Furthermore, we proposed the Geometric Affine Modulation Module (GAM), which uses original modal features to apply affine transformations on initial fusion features to maintain the feature encoder modal discriminability. Finally, we apply a Text-Guided Channel Perturbation Module (TCPM) during decoding to reshape the channel distribution, reducing the dependence on modality-specific channels. Extensive experiments demonstrate that the proposed algorithm outperforms existing methods on both multi-modality image fusion and downstream tasks.

Xiaoyu Li, Peidong Li, Xian Wu, Long Shi, Dedong Liu, Yitao Wu, Jiajia Fu, Dixiao Cui, Lijun Zhao, Lining Sun

Spatio-temporal alignment is crucial for temporal modeling of end-to-end (E2E) perception in autonomous driving (AD), providing valuable structural and textural prior information. Existing methods typically rely on the attention mechanism to align objects across frames, simplifying the motion model with a unified explicit physical model (constant velocity, etc.). These approaches prefer semantic features for implicit alignment, challenging the importance of explicit motion modeling in the traditional perception paradigm. However, variations in motion states and object features across categories and frames render this alignment suboptimal. To address this, we propose HAT, a spatio-temporal alignment module that allows each object to adaptively decode the optimal alignment proposal from multiple hypotheses without direct supervision. Specifically, HAT first utilizes multiple explicit motion models to generate spatial anchors and motion-aware feature proposals for historical instances. It then performs multi-hypothesis decoding by incorporating semantic and motion cues embedded in cached object queries, ultimately providing the optimal alignment proposal for the target frame. On nuScenes, HAT consistently improves 3D temporal detectors and trackers across diverse baselines. It achieves state-of-the-art tracking results with 46.0% AMOTA on the test set when paired with the DETR3D detector. In an object-centric E2E AD method, HAT enhances perception accuracy (+1.3% mAP, +3.1% AMOTA) and reduces the collision rate by 32%. When semantics are corrupted (nuScenes-C), the enhancement of motion modeling by HAT enables more robust perception and planning in the E2E AD.

Xiaopan Li, Yi Jiang, Shiqian Wu, Shoulie Xie, Sos Agaian

Defocus blur, common in shallow depth-of-field photography, varies across image regions and is challenging to accurately estimate and restore. Existing deblurring methods often struggle to capture fine structural textures and do not effectively adapt to regional differences in blur. We propose Multi-Level Blur-Aware Stable Diffusion (MBSD), a novel framework that explicitly integrates regional blur recognition into a diffusion-based image restoration process. MBSD assigns blur-level labels to image patches using a Patch Blur Annotator (PBA), guiding a Multi-Scale Blur Estimator (MSBE) to predict soft blur probabilities and generate routing weights. These weights control a Blur-Adaptive Expert Mixer (BAEM), which adaptively combines features based on local blur severity. The features are then passed to a text-to-image diffusion model via a cross-attention mechanism, enabling region-specific restoration. Extensive experiments on public benchmarks demonstrate that MBSD delivers superior perceptual quality while maintaining competitive PSNR and SSIM, consistently outperforming state-of-the-art methods.

Xianwei Li, Huiyuan Fu, Chuanming Wang, Huadong Ma

With the growing prevalence of HDR-capable cinema venues such as Cinity LED theaters, there is an increasing demand to convert existing Standard Dynamic Range (SDR) films into High Dynamic Range (HDR) formats for theatrical presentation. However, existing SDR-to-HDR conversion methods are primarily tailored for consumer-grade content such as television and therefore fall short of the stringent requirements of professional cinematic material. To bridge this gap, we present HDRMovie7K, the first large-scale, lossless dataset of cinematic SDR-HDR frame pairs sourced from professional Digital Cinema Distribution Master (DCDM) workflows. Based on this foundation, we introduce HDRMovieformer, a transformer-based framework featuring a Luminance Estimator module for luminance guidance, a Luminance-Guided Multi-Head Self-Attention to focus on critical fine-detail recovery, and a Chroma Refiner for color accuracy, optimized with a novel Wide Color Gamut Loss. To further evaluate our model in online streaming media scenarios, we introduce HDRMovie1K, a dataset curated from publicly available HDR film clips. Extensive experiments on both HDRMovie7K and HDRMovie1K demonstrate that our method achieves state-of-the-art performance.

Xiang Li, Ya-Li Li, Yuan Wang, Huaqiang Wang, Shengjin Wang

Recent advances in vision-language-action (VLA) models have demonstrated impressive generalization for robotic manipulation. However, these models often operate by directly mapping visual and linguistic inputs to subsequent actions, lacking intermediate task planning, along with failure detection and recovery ability. These limitations prevent them from effectively decomposing complex tasks, recognizing problems, and correcting erroneous actions, ultimately resulting in complete task failure. This significantly hinders their ability to perform long-horizon tasks and generalization ability. To this end, we introduce TCoT: Trajectory Chain-of-Thought, a unified VLA framework that enhances this direct mapping with trajectory planning as well as failure detection and recovery. TCoT leverages hierarchy trajectories as a precise and compact representation of CoT reasoning for manipulation: global planning provides a high-level, goal-oriented trajectory to guide the robot toward its task objective, while local planning focuses on real-time adjustments to address dynamic changes. Moreover, we designed the Global-Local Switching Recovery algorithm that detects and effectively recovers from failures. Experimental results reveal that TCoT surpasses the state-of-the-art methods across both real and simulated scenarios and exhibits superior generalization capabilities.

Wenrui Li, Yidan Lu, Yeyu Chai, Rui Zhao, Hengyu Man, Xiaopeng Fan

With the daily influx of 3D data on the internet, text-3D retrieval has gained increasing attention. However, current methods face two major challenges: Hierarchy Representation Collapse (HRC) and Redundancy-Induced Saliency Dilution (RISD). HRC compresses abstract-to-specific and whole-to-part hierarchies in Euclidean embeddings, while RISD averages noisy fragments, obscuring critical semantic cues and diminishing the model’s ability to distinguish hard negatives. To address these challenges, we introduce the Hyperbolic Hierarchical Alignment Reasoning Network (H2ARN) for text-3D retrieval. H2ARN embeds both text and 3D data in a Lorentz-model hyperbolic space, where exponential volume growth inherently preserves hierarchical distances. A hierarchical ordering loss constructs a shrinking entailment cone around each text vector, ensuring that the matched 3D instance falls within the cone, while an instance-level contrastive loss jointly enforces separation from non-matching samples. To tackle RISD, we propose a contribution-aware hyperbolic aggregation module that leverages Lorentzian distance to assess the relevance of each local feature and applies contribution-weighted aggregation guided by hyperbolic geometry, enhancing discriminative regions while suppressing redundancy without additional supervision. We also release the expanded T3DR-HIT v2 benchmark, which contains 8,935 text-to-3D pairs, 2.6 times the original size, covering both fine-grained cultural artefacts and complex indoor scenes.

Wenjie Li, Jinglei Shi, Jin Han, Heng Guo, Zhanyu Ma

Clean images are crucial for visual tasks such as small object detection, especially at high resolutions. However, real-world images are often degraded by adverse weather, and weather restoration methods may sacrifice high-frequency details critical for analyzing small objects. A natural solution is to apply super-resolution (SR) after weather removal to recover both clarity and fine structures. However, simply cascading restoration and SR struggle to bridge their inherent conflict: removal aims to remove high-frequency weather-induced noise, while SR aims to hallucinate high-frequency textures from existing details, leading to inconsistent restoration contents. In this paper, we take deraining as a case study and propose DHGM, a Diffusion-based High-frequency Guided Model for generating clean and high-resolution images. DHGM integrates pre-trained diffusion priors with high-pass filters to simultaneously remove rain artifacts and enhance structural details. Extensive experiments demonstrate that DHGM achieves superior performance over existing methods, with lower costs.

Weiran Li, Yeqiang Liu, Qiannan Guo, Yijie Wei, Hwa Liang Leo, Zhenbo Li

Multiple object tracking (MOT) technology has made significant progress in terrestrial applications, but underwater tracking scenarios remain underexplored despite their importance to marine ecology and aquaculture. In this paper, we present Multiple Fish Tracking Dataset 2025 (MFT25), a comprehensive dataset specifically designed for underwater multiple fish tracking, featuring 15 diverse video sequences with 408,578 meticulously annotated bounding boxes across 48,066 frames. Our dataset captures various underwater environments, fish species, and challenging conditions including occlusions, similar appearances, and erratic motion patterns. Additionally, we introduce Scale-aware and Unscented Tracker (SU-T), a specialized tracking framework featuring an Unscented Kalman Filter (UKF) optimized for non-linear swimming patterns of fish and a novel Fish-Intersection-over-Union (FishIoU) matching that accounts for the unique morphological characteristics of aquatic species. Extensive experiments demonstrate that our SU-T baseline achieves state-of-the-art performance on MFT25, with 34.1 HOTA and 44.6 IDF1, while revealing fundamental differences between fish tracking and terrestrial object tracking scenarios.

Wei Li, Long Ji, Ying Wang, Xiao Wu, Zhaoquan Yuan, Penglin Dai

Accurate reconstruction of 3D vehicle pose and shape from monocular images is challenging, particularly for distant objects in autonomous driving. Existing methods often suffer from geometric ambiguity in depth estimation and structural hollowness in shape recovery, primarily due to inadequate multi-scale feature aggregation and unflexible prior modeling. To overcome these limitations, MonoVPR is proposed, a novel framework integrating dynamic context adaptation and progressive geometry refinement. Specifically, a Hierarchical Dual-Context Attention (HDCA) module is introduced to resolve scale-dependent degradation through gated cross-attention across multi-resolution feature maps, dynamically fusing object-centric geometric cues with scene-centric semantics. For shape refinement, the Bounded Iterative Mesh Refiner (BIMR) progressively optimizes template-guided deformations via multi-head attention and a tanh-bounded correction loop, ensuring physically plausible reconstructions.Extensive experiments on the ApolloCar3D benchmark demonstrate MonoVPR achieves state-of-the-art performance, showing exceptional capability in reconstructing geometrically consistent shapes and precise poses for challenging long-range scenarios.

Tiantian Li, Xinjie Zhang, Xingtong Ge, Tongda Xu, Dailan He, Jun Zhang, Yan Wang

Implicit neural representations (INRs) have achieved remarkable success in image representation and compression, but they require substantial training time and memory. Meanwhile, recent 2D Gaussian Splatting (GS) methods (\textit{e.g.}, GaussianImage) offer promising alternatives through efficient primitive-based rendering. However, these methods require excessive Gaussian primitives to maintain high visual fidelity. To exploit the potential of GS-based approaches, we present GaussianImage++, which utilizes limited Gaussian primitives to achieve impressive representation and compression performance. Firstly, we introduce a distortion-driven densification mechanism. It progressively allocates Gaussian primitives according to signal intensity. Secondly, we employ context-aware Gaussian filters for each primitive, which assist in the densification to optimize Gaussian primitives based on varying image content. Thirdly, we integrate attribute-separated learnable scalar quantizers and quantization-aware training, enabling efficient compression of primitive attributes. Experimental results demonstrate the effectiveness of our method. In particular, GaussianImage++ outperforms GaussianImage and INRs-based COIN in representation and compression performance while maintaining real-time decoding and low memory usage.

Sikuang Li, Chen Yang, Jiemin Fang, Taoran Yi, Jia Lu, Jiazhong Cen, Lingxi Xie, Wei Shen, Qi Tian

We tackle the challenge of generating the infinitely extendable 3D world -- large, continuous environments with coherent geometry and realistic appearance. Existing methods face key challenges: 2D-lifting approaches suffer from geometric and appearance inconsistencies across views, 3D implicit representations are hard to scale up, and current 3D foundation models are mostly object-centric, limiting their applicability to scene-level generation. Our key insight is leveraging strong generation priors from pre-trained 3D models for structured scene block generation. To this end, we propose WorldGrow, a hierarchical framework for unbounded 3D scene synthesis. Our method features three core components: (1) a data curation pipeline that extracts high-quality scene blocks for training, making the 3D structured latent representations suitable for scene generation; (2) a 3D block inpainting mechanism that enables context-aware scene extension; and (3) a coarse-to-fine generation strategy that ensures both global layout plausibility and local geometric/textural fidelity. Evaluated on the large-scale 3D-FRONT dataset, WorldGrow achieves SOTA performance in geometry reconstruction, while uniquely supporting infinite scene generation with photorealistic and structurally consistent outputs. These results highlight its capability for constructing large-scale virtual environments and potential for building future world models.

Shuwei Li, Lei Tan, Robby T. Tan

Day-to-night unpaired image translation is important to downstream tasks but remains challenging due to large appearance shifts and the lack of direct pixel-level supervision. Existing methods often introduce semantic hallucinations, where objects from target classes such as traffic signs and vehicles, as well as man-made light effects, are incorrectly synthesized. These hallucinations significantly degrade downstream performance. We propose a novel framework that detects and suppresses hallucinations of target-class features during unpaired translation. To detect hallucination, we design a dual-head discriminator that additionly performs semantic segmentation to identify hallucinated content in background regions. To suppress these hallucinations, we introduce class-specific prototypes, constructed by aggregating features of annotated target-domain objects, which act as semantic anchors for each class. Built upon a Schrödinger Bridge-based translation model, our framework performs iterative refinement, where detected hallucination features are explicitly pushed away from class prototypes in feature space, thus preserving object semantics across the translation trajectory. Experiments show that our method outperforms existing approaches both qualitatively and quantitatively. On the BDD100K dataset, it improves mAP by 15.5% for day-to-night domain adaptation, with a notable 31.7% gain for classes such as traffic lights that are prone to hallucinations.

Shuangzhi Li, Junlong Shen, Lei Ma, Xingyu Li

LiDAR-based 3D object detection models often struggle to generalize to real-world environments due to limited object diversity in existing datasets. To tackle it, we introduce the first generalized cross-domain few-shot (GCFS) task in 3D object detection, aiming to adapt a source-pretrained model to both common and novel classes in a new domain with only few-shot annotations. We propose a unified framework that learns stable target semantics under limited supervision by bridging 2D open-set semantics with 3D spatial reasoning. Specifically, an image-guided multi-modal fusion injects transferable 2D semantic cues into the 3D pipeline via vision-language models, while a physically-aware box search enhances 2D-to-3D alignment via LiDAR priors. To capture class-specific semantics from sparse data, we further introduce contrastive-enhanced prototype learning, which encodes few-shot instances into discriminative semantic anchors and stabilizes representation learning. Extensive experiments on GCFS benchmarks demonstrate the effectiveness and generality of our approach in realistic deployment settings.

Shuaibo Li, Yijun Yang, Zhaohu Xing, Hongqiu Wang, Pengfei Hao, Xingyu Li, Zekai Liu, Qing Zhang, Lei Zhu

The rapid advancement of generative models, which produce increasingly realistic synthetic images, urgently demands robust and generalizable detection methods. Consequently, research has largely pivoted to leveraging large-scale Vision Foundation Models (VFMs) for enhanced generalization. However, existing VFM-based approaches primarily adhere to either perceptual or generative paradigms, each with limitations: perceptual models capture high-level semantics but often miss subtle artifacts, whereas generative models emphasize fine-grained flaws yet overlook semantic inconsistency. To resolve this inherent trade-off, we introduce SynerDetect, a novel hierarchical synergistic framework that fundamentally unifies the two paradigms. SynerDetect achieves deep integration of heterogeneous forensic representations through two levels of synergy: Cross-Model Interactive Distillation (CMID) distills generative forensic signals into perceptual encoders via prompt-guided reconstruction; and Optimal Transport-Guided Discriminative Contrastive Learning (OT-DCL) structurally aligns and integrates these heterogeneous representations, consolidating them into a robust, unified detection space. SynerDetect achieves superior performance on standard benchmarks (AIGCDetectBenchmark and GenImage) and attains a notable 5.20% accuracy gain on the challenging Chameleon benchmark, whose synthetic images consistently pass the Visual Turing Test. These results unequivocally validate the robust, real-world generalization of our unified cross-paradigm framework.

Shiying Li, Xingqun Qi, Bingkun Yang, Weile Chen, Zezhao Tian, Muyi Sun, Qifeng Liu, Man Zhang, Zhenan Sun

Generating responsive listener head dynamics with nuanced emotions and expressive reactions is crucial for dialogue modeling in various virtual avatar animations. Previous studies mainly focus on the direct short-term production of listener behavior. They overlook the fine-grained control over motion variations and emotional intensity, especially in long-sequence modeling. Moreover, the lack of long-term and large-scale paired speaker-listener corpora incorporating head dynamics and fine-grained multi-modality annotations limits the application of dialogue modeling. Therefore, we first newly collect a large-scale multi-turn dataset of 3D dyadic conversation containing more than 1.4M valid frames for multi-modal responsive interaction, dubbed ListenerX. Additionally, we propose VividListener, a novel framework enabling fine-grained, expressive, and controllable listener dynamics modeling. This framework leverages multi-modal conditions as guiding principles for fostering coherent interactions between speakers and listeners. Specifically, we design the Responsive Interaction Module (RIM) to adaptively represent the multi-modal interactive embeddings. RIM ensures the listener dynamics achieve fine-grained semantic coordination with textual descriptions and adjustments, while preserving expressive reaction with speaker behavior. Meanwhile, we propose the Emotional Intensity Tags (EIT) for emotion intensity editing with multi-modal information integration, applying to both text descriptions and listener motion amplitude. Extensive experiments conducted on our newly collected ListenerX dataset demonstrate that VividListener achieves state-of-the-art performance, realizing expressive and controllable listener dynamics.

Shijie Li, Yiming Chen, Zhineng Chen, Kai Hu, Xieping Gao

Accurate segmentation of ultra-high-resolution (UHR) images, which often exceed tens of millions of pixels, is critically important in domains such as remote sensing and biomedical imaging. However, acquiring pixel-level annotations for such high-resolution images is prohibitively expensive and labor-intensive. While semi-supervised semantic segmentation can significantly reduce the annotation burden, its extension to UHR images holds great potential for addressing the unique challenges posed by sparse supervision. To this end, we propose SSR-SAM, a retrieval-style semi-supervised segmentation framework tailored for UHR images. Leveraging the promptable paradigm of the Segment Anything Model (SAM), SSR-SAM treats locally annotated regions as prompts to retrieve semantically consistent pixels across the entire image. Building upon this retrieval-style segmentation paradigm, we further introduce prompt-level perturbation, a novel trail to deploy consistency regularization for semi-supervised segmentation. It encourages the model to learn consistency across predictions guided by diverse visual-semantic prompts, thereby enhancing generalization on unlabeled data. We evaluate SSR-SAM on three UHR datasets: Inria Aerial, BCSS, and URUR. Experimental results show that SSR-SAM achieves clear performance gains over the labeled-only supervision, with average mIoU improvements of 4.9%, 4.15%, and 2.5%, respectively. Additionally, SSR-SAM possesses zero-shot segmentation capability, exhibiting potential for general retrieval-style segmentation tasks.

Shicheng Li, Lei Li, Kun Ouyang, Shuhuai Ren, Yuanxin Liu, Yuanxing Zhang, Fuzheng Zhang, Lingpeng Kong, Qi Liu, Xu Sun

Video Large Language Models (Video LLMs) have achieved significant success by adopting the paradigm of large-scale pre-training followed by supervised fine-tuning (SFT). However, existing approaches struggle with temporal reasoning due to weak temporal correspondence in the data and over-reliance on the next-token prediction paradigm, which collectively result in the absence temporal supervision. To address these limitations, we propose TEMPLE (TEMporal Preference Learning), a systematic framework that enhances temporal reasoning capabilities through Direct Preference Optimization (DPO). To address temporal information scarcity in data, we introduce an automated pipeline for systematically constructing temporality-intensive preference pairs comprising three steps: selecting temporally rich videos, designing video-specific perturbation strategies, and evaluating model responses on clean and perturbed inputs. Complementing this data pipeline, we provide additional supervision signals via preference learning and propose a novel Progressive Pre-SFT Alignment strategy featuring two key innovations: a curriculum learning strategy which progressively increases perturbation difficulty to maximize data efficiency; and applying preference optimization before instruction tuning to incentivize fundamental temporal alignment. Extensive experiments demonstrate that our approach consistently improves Video LLM performance across multiple benchmarks with a relatively small set of self-generated DPO data. Our findings highlight TEMPLE as a scalable and efficient complement to SFT-based methods, paving the way for developing reliable Video LLMs.

Shaoxu Li, Ye Pan

Despite significant advancements in image generation using advanced generative frameworks, cross-image integration of content and style remains a key challenge. Current generative models, while powerful, frequently depend on vague textual prompts to define styles—creating difficulties in balancing content semantics and style preservation. We propose a novel framework that utilizes customized models to learn style representations. It enhances content preservation through cross-model feature and attention modulation, leveraging the inherent semantic consistency across models. Additionally, we introduce fixed feature and adaptive attention fusion to achieve the desired balance between content and style. We further develop spatial (mask-guided localized) and temporal (multi-style compositional) multi-model combinations, enabling flexible fusion of models and styles. Extensive experiments demonstrate that our method outperforms state-of-the-art approaches in balancing content preservation and stylistic coherence.

Ruibo Li, Xiaofeng Yang, Ze Yang, Jiacheng Wei, Chunyan Miao, Guosheng Lin

LiDAR data generation has emerged as a promising solution to the high cost and limited scalability of real-world LiDAR sensing. Recent diffusion and rectified flow models have demonstrated strong capabilities in synthesizing realistic 3D point clouds; however, their iterative sampling procedures result in significant inference overhead. To address this, we focus on efficient few-step LiDAR generation for both unconditional and multi-modal conditional settings. Specifically, we propose an adaptive piecewise distillation strategy tailored for rectified flow-based LiDAR generation models, where the teacher model’s flow trajectory is adaptively segmented into consecutive intervals, and the student is trained only at the start of each interval to directly predict the velocity toward its endpoint. By sequentially sampling at the start timestep of each interval, our method enables fast few-step generation. Moreover, instead of uniform partitioning, we introduce an adaptive timestep selection strategy that chooses interval boundaries with minimal initial error, thereby reducing the complexity of distillation. Experimental results show that our method achieves comparable or superior performance to state-of-the-art methods in both unconditional and multi-modal conditional LiDAR generation, using only four sampling steps.