论文检索

Yu Qi, Hongyu Li, Shaofei Huang, Tianrui Hui, Yaxiong Wang, Lechao Cheng, Zhun Zhong, Si Liu, Meng Wang

In this paper, we tackle the Aerial Vision-and-Dialog Navigation (AVDN) task in the training-free setting for resource-efficient high-altitude UAV navigation.Naively applying MLLMs leads to unreliable navigation due to weak directional grounding and the lack of explicit spatial memory.To address these issues, we propose PSC-AVDN, a training-free framework that tightly couples a three-stage Parsing-Search-Confirmation reasoning pipeline with a Structured Spatial Memory (SSM).The parsing stage uses an LLM to convert ambiguous dialogue instructions into stable geometric directional and destination cues.A Search Chain-of-Thought (S-CoT) then performs stepwise target exploration under high-altitude observations, and a Confirmation Chain-of-Thought (C-CoT) conducts fine-grained verification around candidate regions to resolve visual ambiguity.Meanwhile, SSM integrates three complementary sources of spatial cues, including multi-scale visual observation, spatial visual memory, and structured geometric memory to provide global spatial context and long-horizon consistency.Extensive experiments on ANDH and ANDH-Full show that PSC-AVDN establishes new state-of-the-art performance in the training-free setting, matching or surpassing several finetuned methods.

Yahong Wang, Juncheng Wu, Zhangkai Ni, Longzhen Yang, Yihang Liu, Chengmei Yang, Ying Wen, Lianghua He, Xianfeng Tang, Hui Liu 等

Vision Large Language Models (VLLMs) incur high computational costs due to their reliance on hundreds of visual tokens to represent images. While token pruning offers a promising solution for accelerating inference, this paper, however, identifies a key observation: in deeper layers (e.g., beyond the 20th), existing training-free pruning methods perform no better than random pruning. We hypothesize that this degradation is caused by "vanishing token information", where visual tokens progressively lose their salience with increasing network depth. To validate this hypothesis, we quantify a token's information content by measuring the change in the model output probabilities upon its removal. Using this proposed metric, our analysis of the information of visual tokens across layers reveals three key findings: (1) As layers deepen, the information of visual tokens gradually becomes uniform and eventually vanishes at an intermediate layer, which we term as "information horizon", beyond which the visual tokens become redundant;(2) The position of this horizon is not static; it extends deeper for visually intensive tasks, such as Optical Character Recognition (OCR), compared to more general tasks like Visual Question Answering (VQA);(3) This horizon is also strongly correlated with model capacity, as stronger VLLMs (e.g., Qwen2.5-VL) employ deeper visual tokens than weaker models (e.g., LLaVA-1.5).Based on our findings, we show that simple random pruning in deep layers efficiently balances performance and efficiency. Moreover, integrating random pruning consistently enhances existing methods. Using DivPrune with random pruning achieves state-of-the-art results, maintaining 96.9% of Qwen-2.5-VL-7B performance while pruning 50% of visual tokens. The code is available at https://github.com/YahongWang1/Information-Horizon.

Junsung Lee, Hyunsoo Lee, Yong Jae Lee, Bohyung Han

High-resolution content creation is rapidly emerging as a central challenge in both the vision and graphics communities. Images serve as the most fundamental modality for visual expression, and content generation that aligns with the user intent requires effective, controllable high-resolution image manipulation mechanisms. However, existing approaches remain limited to low-resolution settings, typically supporting only up to 1K resolution. In this work, we introduce the task of high-resolution image editing and propose a test-time optimization framework to address it. Our method performs patch-wise optimization on high-resolution source images, followed by a fine-grained detail transfer module and a novel synchronization strategy to maintain consistency across patches. Extensive experiments show that our method produces high-quality edits, facilitating high-resolution content creation.

Sheldon Fung, Wei Pan, Ling Cao, Fei Hou, Ling Chen, Shasha Mao, Hongdong Li, Xuequan Lu

While existing point-cloud-registration methods can well handle high-overlap scenarios of two point clouds, they often struggle with low-overlap scenarios, due to inevitable geometric/semantic ambiguities in the non-overlapping regions. In this paper, we introduce SuP, a novel framework that reformulates low-overlap registration as a high-overlap sub-cloud pairs (anchor pairs) mining problem. Central to SuP is our Dual-phase Sub-cloud Anchor Mining (DSAM) module, which first subdivides the source and target point clouds into multiple sub-clouds, followed by introducing a dual-phase weighting pipeline: 1) an efficient overlap-guided prior-weighting scheme (OPS) that leverages feature salience to identify candidate anchor pairs, and 2) a multi-scale post-weighting network (MPN) that exploits neighborhood feature consensus to further identify anchor pairs. Subsequently, final correspondences are generated through a merge-to-match module using the anchor pairs. To train DSAM, we design an alignment-aware weighting loss that uses on-the-fly alignment errors as supervision. Comprehensive experiments on the color-enhanced 3DMatch and 3DLoMatch demonstrate that SuP significantly outperforms state-of-the-art methods, achieving higher registration recall and more accurate alignment, especially under challenging low-overlap conditions.

Qinbo Zhang, Yanhang Shi, Ziyi Zhang, Hao Wang, Sai Qian Zhang, Jian Li

Vertical federated learning (VFL) trains models by splitting computation across clients and a server that only exchange intermediate embeddings. Recent work shows that a server even if honest-but-curious can steal a client's bottom model by querying the system and regressing on the returned embeddings, and in response, defenses perturb or decouple the embedding channel. We show these defenses remain vulnerable. We propose VENOM, a geometry-aware stealing attack. VENOM first learns a contrastive space over server-observed embeddings, then builds a neighborhood graph and trains a surrogate bottom model to match targets and respect local geometry via a neighbor-matching loss alongside pointwise and feature-shape alignment. This strategy preserves the relational structure that defenses fail to erase, effectively recoupling the embeddings produced by multi-branch and noise-based defenses. Across six datasets, VENOM consistently outperforms standard stealing methods under no defense and multiple defenses, and remains effective with out-of-distribution (OOD) auxiliary data.

Albert Dominguez Mantes, Gioele La Manno, Martin Weigert

Modern microscopy routinely produces gigapixel images that contain structures across multiple spatial scales, from fine cellular morphology to broader tissue organization. Many analysis tasks require combining these scales, yet most vision models operate at a single resolution or derive multi-scale features from one view, limiting their ability to exploit the inherently multi-resolution nature of microscopy data. We introduce MuViT, a transformer architecture built to fuse true multi-resolution observations from the same underlying image. MuViT embeds all patches into a shared world-coordinate system and extends rotary positional embeddings to these coordinates, enabling attention to integrate large-scale context with high-resolution detail within a single encoder. Across synthetic benchmarks, kidney histopathology, and high-resolution mouse-brain microscopy, MuViT delivers consistent improvements over strong ViT and CNN baselines. Multi-resolution MAE pretraining further produces scale-consistent representations that enhance downstream tasks. These results demonstrate that explicit world-coordinate modelling provides a simple yet powerful mechanism for leveraging multi-resolution information in large-scale microscopy analysis. Code is available at github.com/weigertlab/muvit.

Nikolay Nikolov, Giuliano Albanese, Sombit Dey, Aleksandar Yanev, Luc Van Gool, Jan-Nico Zaech, Danda Pani Paudel

Robotic Foundation Models (RFMs) hold great promise as generalist, end-to-end systems for robot control.Yet their ability to generalize across new environments, tasks, and embodiments remains limited.We argue that a major bottleneck lies in their foundations: most RFMs are built by fine-tuning internet-pretrained Vision-Language Models (VLMs).However, these VLMs are trained on 2D image-language tasks and lack the 3D spatial reasoning inherently required for embodied control in the 3D world.Bridging this gap directly with large-scale robotic data is costly and difficult to scale.Instead, we propose to enrich easy-to-collect non-robotic image data with 3D annotations and enhance a pretrained VLM with 3D understanding capabilities.Following this strategy, we train SPEAR-VLM, a 3D-aware VLM that infers object coordinates in 3D space from a single 2D image.Building on SPEAR-VLM, we introduce our main contribution, SPEAR-1: a robotic foundation model that integrates grounded 3D perception with language-instructed embodied control.Trained on ~45M frames from 24 Open X-Embodiment datasets, SPEAR-1 outperforms or matches state-of-the-art models such as \pi_0-FAST and \pi_ 0.5 , while it uses 20xfewer robot demonstrations.This carefully-engineered training strategy unlocks new VLM capabilities and as a consequence boosts the reliability of embodied control beyond what is achievable with only robotic data.We make our model weights and 3D-annotated datasets publicly available.

Zhiqiang Kou, Junxiang Wu, Wenke Huang, Wenwen He, Ming-Kun Xie, Changwei Wang, Yuheng Jia, Di Jiang, Yang Liu, Xin Geng 等

Federated Multi-Label Learning is a distributed paradigm where multiple clients possess heterogeneous multi-label data and perform collaborative learning under privacy constraints without sharing raw data. However, modeling label correlations under heterogeneous distributions remains challenging. Due to client-specific label spaces and varying co-occurrence patterns, correlations learned by individual clients inevitably deviate from the global structure, a phenomenon we term label correlation drift. To address this, we propose FedHarmony, a framework that harmonizes heterogeneous label correlations across clients. It introduces consensus correlation, capturing agreement among clients and serving as a global teacher to correct biased local estimates. During aggregation, FedHarmony evaluates each client by both data size and correlation quality, assigning weights accordingly. Moreover, we develop an accelerated optimization algorithm and theoretically establish faster convergence without sacrificing accuracy. Experiments on real-world federated multi-label datasets show that FedHarmony consistently outperforms state-of-the-art methods.

Taehoon Kim, Henry Gouk, Timothy Hospedales

Test-time alignment (TTA) aims to adapt models to specific rewards during inference. However, existing methods tend to either under-optimise or over-optimise (reward hack) the target reward function. We propose Null-Text Test-Time Alignment (Null-TTA), which aligns diffusion models by optimising the unconditional embedding in classifier-free guidance, rather than manipulating latent or noise variables. Due to the structured semantic nature of the text embedding space, this ensures alignment occurs on a semantically coherent manifold and prevents reward hacking (exploiting non-semantic noise patterns to improve the reward). Since the unconditional embedding in classifier-free guidance serves as the anchor for the model's generative distribution, Null-TTA directly steers model's generative distribution towards the target reward rather than just adjusting the samples, even without updating model parameters. Thanks to these desirable properties, we show that Null-TTA achieves state-of-the-art target test-time alignment while maintaining strong cross-reward generalisation. This establishes semantic-space optimisation as an effective and principled novel paradigm for TTA.

Yuwen Tan, Yuan Qing, Boqing Gong

This paper reveals that many open-source large language models (LLMs) lack hierarchical knowledge about our visual world, unaware of even well-established biology taxonomies. This shortcoming makes LLMs a bottleneck for vision LLMs' hierarchical visual recognition (e.g., recognizing Anemone Fish but not Vertebrate). We arrive at these findings using about one million four-choice visual question answering (VQA) tasks constructed from six taxonomies and four image datasets. Interestingly, finetuning a vision LLM using our VQA tasks reaffirms LLMs' bottleneck effect because the VQA tasks improve the LLMs' hierarchical consistency in text-only tasks more than the vision LLMs'. We believe that one cannot make vision LLMs understand our visual world hierarchically until LLMs possess corresponding taxonomy knowledge.

Yongwei Jiang, Yixiong Zou, Yuhua Li, Ruixuan Li

Cross-domain few-shot object detection (CD-FSOD) aims to adapt pretrained detectors from a source domain to target domains with limited annotations, suffering from severe domain shifts and data scarcity problems. In this work, we find a previously overlooked phenomenon: models exhibit dispersed and unfocused attention in target domains, leading to imprecise localization and redundant predictions, just like a human cannot focus on visual objects. Therefore, we call it the target-domain Astigmatism problem. Analysis on attention distances across transformer layers reveals that regular fine-tuning inherently shows a trend to remedy this problem, but results are still far from satisfactory, which we aim to enhance in this paper. Biologically inspired by the human fovea-style visual system, we enhance the fine-tuning's inherent trend through a center-periphery attention refinement framework, which contains (1) a Positive Pattern Refinement module to reshape attention toward semantic objects using class-specific prototypes, simulating the visual center region; (2) a Negative Context Modulation module to enhance boundary discrimination by modeling background context, simulating the visual periphery region; and (3) a Textual Semantic Alignment module to strengthen center-periphery distinction through cross-modal cues. Our bio-inspired approach transforms astigmatic attention into focused patterns, substantially improving adaptation to target domains. Experiments on six challenging CD-FSOD benchmarks consistently demonstrate improved detection accuracy and establish new state-of-the-art results.

Sijie Li, Biao Qian, Jungong Han

Network pruning is an effective technique for enabling lightweight Large Vision-Language Models (LVLMs), which primarily incorporates both weights and activations into the importance metric. However, existing efforts typically process calibration data from different modalities in a unified manner, overlooking modality-specific behaviors. This raises a critical challenge: how to address the divergent behaviors of textual and visual tokens for accurate pruning of LVLMs. To this end, we systematically investigate the sensitivity of visual and textual tokens to the pruning operation by decoupling their corresponding weights, revealing that: (i) the textual pathway should be calibrated via text tokens, since it exhibits higher sensitivity than the visual pathway; (ii) the visual pathway exhibits high redundancy, permitting even 50% sparsity. Motivated by these insights, we propose a simple yet effective Asymmetric Text-Visual Weight Pruning method for LVLMs, dubbed ATV-Pruning, which establishes the importance metric for accurate weight pruning by selecting the informative tokens from both textual and visual pathways. Specifically, ATV-Pruning integrates two primary innovations: first, a calibration pool is adaptively constructed by drawing on all textual tokens and a subset of visual tokens; second, we devise a layer-adaptive selection strategy to yield important visual tokens. Finally, extensive experiments across standard multimodal benchmarks verify the superiority of our ATV-Pruning over state-of-the-art methods.

Yuanpeng Tu, Yunpeng Chen, Xinyu Zhang, Chao Liao, Hengshuang Zhao

MeanFlow is a powerful few-step generative framework that can be trained from scratch, but its performance degrades significantly when the one-step loss uses a large portion of training data. This stems from a temporal scale imbalance: gradients from different stages of generation contribute unevenly, leading to unstable optimization--evident in blurry samples and high FID scores. The core issue is a conflict between two opposing forces: terms that amplify variance over long time spans and strong constraints needed near the start of generation, which a fixed sampling strategy cannot reconcile. To resolve this, we propose Temporal Equilibrium MeanFlow (TEMF), which balances these competing demands through two simple yet effective components: (1) a temporal equilibrium weighting function that equalizes gradient influence across all time scales, and (2) a dynamic boundary scheduler that gradually shifts training focus--from stabilizing early steps to refining the full trajectory as training progresses. Without changing the model architecture, TEMF retains true one-step generation with classifier-free guidance, achieving a state-of-the-art FID of 2.62 on ImageNet 256x256--achieving the best results among diffusion- and flow-based one-step methods.

Yuhang Ma, Bo Cheng, Shanyuan Liu, Hongyi Zhou, Liebucha Wu, Dawei Leng, Yuhui Yin

Flow-based Transformer models have achieved state-of-the-art image generation performance, but often suffer from high inference latency and computational cost due to their large parameter sizes. To improve inference efficiency without compromising quality, we propose Bridged Progressive Rectified Flow Transformers (NAMI), which decompose the generation process across temporal, spatial, and architectural demensions. We divide the rectified flow into different stages according to resolution, and use a BridgeFlow module to connect them. Fewer Transformer layers are used at low-resolution stages to generate image layouts and concept contours, and more layers are progressively added as the resolution increases. Experiments demonstrate that our approach achieves fast convergence and reduces inference time while ensuring generation quality. The main contributions of this paper are summarized as follows: (1) We introduce Bridged Progressive Rectified Flow Transformers that enable multi-resolution training, accelerating model convergence; (2) NAMI leverages piecewise flow and spatial cascading of Diffusion Transformer (DiT) to rapidly generate images, reducing inference time by 64% for generating 1024x1024 resolution images; (3) We propose a BridgeFlow module to align flows between different stages; (4) We propose the NAMI-1K benchmark to evaluate human preference performance, aiming to mitigate distributional bias and comprehensively assess model effectiveness. The results show that our model is competitive with state-of-the-art models.

Bingyi Cao, Koert Chen, Kevis-Kokitsi Maninis, Kaifeng Chen, Arjun Karpur, Ye Xia, Sahil Dua, Tanmaya Dabral, Guangxing Han, Bohyung Han 等

Recent progress in vision-language pretraining has enabled significant improvements to many downstream computer vision applications, such as classification, retrieval, segmentation and depth prediction. However, a fundamental capability that these models still struggle with is aligning dense patch representations with text embeddings of corresponding concepts. In this work, we investigate this critical issue and propose novel techniques to enhance this capability in foundational vision-language models. First, we reveal that a patch-level distillation procedure significantly boosts dense patch-text alignment -- surprisingly, the patch-text alignment of the distilled student model strongly surpasses that of the teacher model. This observation inspires us to consider modifications to pretraining recipes, leading us to propose iBOT++, an upgrade to the commonly-used iBOT masked image objective, where unmasked tokens also contribute directly to the loss. This dramatically enhances patch-text alignment of pretrained models. Additionally, to improve vision-language pretraining efficiency and effectiveness, we modify the exponential moving average setup in the learning recipe, and introduce a caption sampling strategy to benefit from synthetic captions at different granularities. Combining these components, we develop TIPSv2, a new family of image-text encoder models suitable for a wide range of downstream applications. Through comprehensive experiments on 9 tasks and 20 datasets, we demonstrate strong performance, generally on par with or better than recent vision encoder models. Code and models are released via our project page at https://gdm-tipsv2.github.io/.

Ahyoung Oh, Wonseok Shin, Songkuk Kim

Sparse Autoencoders (SAEs) have demonstrated significant success in interpreting Large Language Models (LLMs) by decomposing dense representations into sparse, semantic components. However, their potential for analyzing Vision Transformers (ViTs) remains largely under-explored. In this work, we present the first application of SAEs to the ViT [CLS] token for out-of-distribution (OOD) detection, addressing the limitation of existing methods that rely on entangled feature representations. We propose a novel framework utilizing a Top-k SAE to disentangle the dense [CLS] features into a structured latent space. Through this analysis, we reveal that in-distribution (ID) data exhibits consistent, class-specific activation patterns, which we formalize as Class Activation Profiles (CAPs). Our study uncovers a key structural invariant: while ID samples preserve a stable pattern within CAPs, OOD samples systematically disrupt this structure. Leveraging this insight, we introduce a scoring function based on the divergence of core energy profiles to quantify the deviation from ideal activation profiles. Our method achieves strong results on the FPR95 metric--critical for safety-sensitive applications--across multiple benchmarks, while also achieving competitive AUROC. Overall, our findings demonstrate that the sparse, disentangled features revealed by SAEs can serve as a powerful, interpretable tool for robust OOD detection in vision models.

Yu Gao, Lutong Su, Ruixiang Huang, Tianji Jiang, Jiadong Tang, Yufeng Yue, Yi Yang

High-quality 3D scene representation in radiance fields relies on accurate camera poses which are often difficult to acquire in real-world scenarios. An effective solution is to use RGB images for the joint optimization of radiance fields and camera poses, an approach that has been well explored in NeRF series methods. However, unlike NeRF, joint optimization in 3D Gaussian Splatting (3DGS) often requires additional regularization or prior spatial knowledge to reach comparable performance. To eliminate these dependencies, we introduce Energy-GS, a pose-aware Gaussian splatting framework that jointly optimizes scene representation and camera poses using only RGB images. We observe that pose gradients in joint optimization are unstable due to the point-based rendering mechanism. Furthermore, unlike NeRF's spatial sampling framework that enables coarse-to-fine pose alignment, rasterization-based 3DGS lacks controllable sampling and thus cannot support progressive pose refinement. To address these challenges, we redesign the optimization strategy of Gaussian primitives and introduce an image-energy-guided constraint that encourages progressive alignment of camera poses. Experiments on both synthetic and real-world datasets show that Energy-GS can effectively optimize the scene reconstruction and resolve camera pose misalignment at the same time. Benefiting from reliance on only RGB images, we believe this work provides promising insights for visual localization and dense mapping applications such as SLAM. The Code is available at \href https://github.com/SkylerGao/ENGS https://github.com/SkylerGao/ENGS

Minxue Tang, Yangyang Yu, Aolin Ding, Maziyar Baran Pouyan, Taha Belkhouja, Yujia Bao

Recognizing visual and textual patterns is essential in many real-world applications of modern AI. However, tackling long-tail pattern recognition tasks remains challenging for current pre-trained foundation models such as LLMs and VLMs. While finetuning pre-trained models can improve accuracy in recognizing such implicit patterns, it is usually infeasible due to a lack of training data and high computational overhead. In this paper, we propose ADAMAB, an efficient embedding calibration framework for few-shot pattern recognition. To maximally reduce the computational costs, ADAMAB trains embedder-agnostic light-weight calibrators on top of fixed embedding models without accessing their parameters. To mitigate the need for large-scale training data, we introduce an adaptive data augmentation strategy based on the Multi-Armed Bandit (MAB) mechanism. With a modified upper confidence bound algorithm, ADAMAB diminishes the gradient shifting and offers theoretically guaranteed convergence in few-shot training. Our multi-modal experiments justify the superior performance of ADAMAB, with up to 40% accuracy improvement when training with less than 5 initial data samples of each class.

Tianchen Guo, Chen Liu, Xin Yu

Human perception of social environments is inherently a multi-view synthesis problem, requiring the integration of complementary and often occluded information across space and time. However, existing benchmarks for Multimodal Large Language Models (MLLMs) are overwhelmingly predicated on a "sufficient-view" assumption, rewarding single-view pattern recognition while failing to evaluate cross-view fusion. To address this critical gap, we introduce CVBench, a large-scale, multi-task benchmark for cross-view human understanding. CVBench comprises 3,000 challenging questions across 12 spatial and temporal tasks, where every item is designed with verifiable single-view insufficiency, mandating that models synthesize disparate evidence to resolve ambiguities. Our comprehensive evaluation of state-of-the-art open and closed-source MLLMs reveals a substantial performance gap, with the best models falling nearly 50 points behind human performance. We identify a systemic failure mechanism across all models: a dominant "Single-View Bias", whereby models ignore conflicting evidence and default to the most confident but incorrect single-view prediction. This demonstrates that current MLLMs lack the fundamental mechanisms for geometric grounding, identity persistence, and true spatio-temporal fusion. CVBench provides a rigorous diagnostic framework to catalyze the development of next-generation, cross-view-aware architectures.

Qiuyang Zhang, Jiujun Cheng, Qichao Mao, Cong Liu, Yu Fang, Yuhong Li, Mengying Ge, Shangce Gao

Spiking Neural Networks (SNNs) promise energy-efficient vision, but applying them to RGB visual tracking remains difficult: Existing SNN tracking frameworks either do not fully align with spike-driven computation or do not fully leverage neurons' spatiotemporal dynamics, leading to a trade-off between efficiency and accuracy. To address this, we introduce SpikeTrack, a spike-driven framework for energy-efficient RGB object tracking. SpikeTrack employs a novel asymmetric design that uses asymmetric timestep expansion and unidirectional information flow, harnessing spatiotemporal dynamics while cutting computation. To ensure effective unidirectional information transfer between branches, we design a memory-retrieval module inspired by neural inference mechanisms. This module recurrently queries a compact memory initialized by the template to retrieve target cues and sharpen target perception over time. Extensive experiments demonstrate that SpikeTrack achieves the state-of-the-art among SNN-based trackers and remains competitive with advanced ANN trackers. Notably, it surpasses TransT on LaSOT dataset while consuming only 1/26 of its energy. To our knowledge, SpikeTrack is the first spike-driven framework to make RGB tracking both accurate and energy efficient.