Peer Revenue Past citations are no guarantee of future returns

Papers

Fresh from arXiv, typeset properly. Highlighted words trade on the market.

Computer Science Computer Vision

cs.CV Watch Computer Vision (sign in to use)

Wednesday 26 November

Computer Vision

DiverseVAR: Balancing Diversity and Quality of Next-Scale Visual Autoregressive Models

We introduce DiverseVAR, a framework that enhances the diversity of text-conditioned visual autoregressive models (VAR) at test time without requiring retraining, fine-tuning, or substantial computational overhead. While VAR models have recently emerged as strong competitors to diffusion and flow models for image generation, they suffer from a critical limitation in diversity, often producing nearly identical images even for simple prompts. This issue has largely gone unnoticed amid the predominant focus on image quality. We address this limitation at test time in two stages. First, inspired by diversity enhancement techniques in diffusion models, we propose injecting noise into the text embedding. This introduces a trade-off between diversity and image quality: as diversity increases, the image quality sharply declines. To preserve quality, we propose scale-travel: a novel latent refinement technique inspired by time-travel strategies in diffusion models. Specifically, we use a multi-scale autoencoder to extract coarse-scale tokens that enable us to resume generation at intermediate stages. Extensive experiments show that combining text-embedding noise injection with our scale-travel refinement significantly enhances diversity while minimizing image-quality degradation, achieving a new Pareto frontier in the diversity-quality trade-off. Project page: https://diverse-var.github.io
Read the full abstract

All authors Mingue Park, Prin Phunyaphibarn, Phillip Y. Lee, Minhyuk Sung

Computer Vision

Understanding the Effects of Distractors on Reasoning Vision-Language Models

How does irrelevant information (i.e., distractors) affect test-time scaling in vision-language models (VLMs)? Prior work on text-only language models has shown that textual distractors can intensify inverse scaling, causing models to reason longer but less effective reasoning traces. In this work, we investigate whether similar phenomena arise in multimodal settings. We introduce Idis (Images with distractors), a visual question-answering dataset that systematically varies distractors along semantic and numerical dimensions. Our analyses reveal that visual distractors affect reasoning VLMs in a fundamentally different way from textual distractors: although inverse scaling still emerges, visual distractors reduce accuracy without increasing reasoning length. We further show that attribute counts extracted from reasoning traces provide key insights into how distractors interact with reasoning length and accuracy. As a sanity check, we propose a simple prompting strategy that mitigates distractor-driven predictions in reasoning vision-language models.
Read the full abstract

All authors Jiyun Bae, Hyunjong Ok, Sangwoo Mo, Jaeho Lee

Computer Vision

Monet: Reasoning in Latent Visual Space Beyond Images and Language

"Thinking with images" has emerged as an effective paradigm for advancing visual reasoning, extending beyond text-only chains of thought by injecting visual evidence into intermediate reasoning steps. However, existing methods fall short of human-like abstract visual thinking, as their flexibility is fundamentally limited by external tools. In this work, we introduce Monet, a training framework that enables multimodal large language models (MLLMs) to reason directly within the latent visual space by generating continuous embeddings that function as intermediate visual thoughts. We identify two core challenges in training MLLMs for latent visual reasoning: high computational cost in latent-vision alignment and insufficient supervision over latent embeddings, and address them with a three-stage distillation-based supervised fine-tuning (SFT) pipeline. We further reveal a limitation of applying GRPO to latent reasoning: it primarily enhances text-based reasoning rather than latent reasoning. To overcome this, we propose VLPO (Visual-latent Policy Optimization), a reinforcement learning method that explicitly incorporates latent embeddings into policy gradient updates. To support SFT, we construct Monet-SFT-125K, a high-quality text-image interleaved CoT dataset containing 125K real-world, chart, OCR, and geometry CoTs. Our model, Monet-7B, shows consistent gains across real-world perception and reasoning benchmarks and exhibits strong out-of-distribution generalization on challenging abstract visual reasoning tasks. We also empirically analyze the role of each training component and discuss our early unsuccessful attempts, providing insights for future developments in visual latent reasoning. Our model, data, and code are available at https://github.com/NOVAglow646/Monet.
Read the full abstract

All authors Qixun Wang, Yang Shi, Yifei Wang, Yuanxing Zhang, Pengfei Wan, Kun Gai, Xianghua Ying, Yisen Wang

Computer Vision

Thinking With Bounding Boxes: Enhancing Spatio-Temporal Video Grounding via Reinforcement Fine-Tuning

Spatio-temporal video grounding (STVG) requires localizing a target object in untrimmed videos both temporally and spatially from natural language descriptions. Despite their strong language understanding, multimodal large language models (MLLMs) underperform on STVG due to misaligned training objectives and weak fine-grained region-word alignment in standard visual encoders. To address this, we propose STVG-o1, the first framework that enables off-the-shelf MLLMs to achieve state-of-the-art STVG performance without any architectural modifications. Our method introduces a bounding-box chain-of-thought mechanism that explicitly reasons about spatio-temporal locations in an intermediate step before producing the final prediction. We further design a multi-dimensional reinforcement reward function consisting of format, consistency, temporal, spatial, and think rewards, which provides geometry-aware supervision through reinforcement fine-tuning. Evaluated on HCSTVG-v1/v2 and VidSTG, STVG-o1 sets new state-of-the-art results on HCSTVG, outperforming the best task-specific method by 7.3% m\_tIoU on HCSTVG-v1, matching specialized models on VidSTG, and surpassing all existing MLLM-based approaches by large margins. It also demonstrates strong open-vocabulary generalization across datasets, establishing MLLMs as viable and powerful backbones for precise spatio-temporal grounding. Our code and models will be released.
Read the full abstract

All authors Xin Gu, Haoji Zhang, Qihang Fan, Jingxuan Niu, Zhipeng Zhang, Libo Zhang, Guang Chen, Fan Chen, Longyin Wen, Sijie Zhu

Computer Vision

Endo-G$^{2}$T: Geometry-Guided & Temporally Aware Time-Embedded 4DGS For Endoscopic Scenes

Endoscopic (endo) video exhibits strong view-dependent effects such as specularities, wet reflections, and occlusions. Pure photometric supervision misaligns with geometry and triggers early geometric drift, where erroneous shapes are reinforced during densification and become hard to correct. We ask how to anchor geometry early for 4D Gaussian splatting (4DGS) while maintaining temporal consistency and efficiency in dynamic endoscopic scenes. Thus, we present Endo-G$^{2}$T, a geometry-guided and temporally aware training scheme for time-embedded 4DGS. First, geo-guided prior distillation converts confidence-gated monocular depth into supervision with scale-invariant depth and depth-gradient losses, using a warm-up-to-cap schedule to inject priors softly and avoid early overfitting. Second, a time-embedded Gaussian field represents dynamics in XYZT with a rotor-like rotation parameterization, yielding temporally coherent geometry with lightweight regularization that favors smooth motion and crisp opacity boundaries. Third, keyframe-constrained streaming improves efficiency and long-horizon stability through keyframe-focused optimization under a max-points budget, while non-keyframes advance with lightweight updates. Across EndoNeRF and StereoMIS-P1 datasets, Endo-G$^{2}$T achieves state-of-the-art results among monocular reconstruction baselines.
Read the full abstract

All authors Yangle Liu, Fengze Li, Kan Liu, Jieming Ma

Computer Vision

PFF-Net: Patch Feature Fitting for Point Cloud Normal Estimation

Estimating the normal of a point requires constructing a local patch to provide center-surrounding context, but determining the appropriate neighborhood size is difficult when dealing with different data or geometries. Existing methods commonly employ various parameter-heavy strategies to extract a full feature description from the input patch. However, they still have difficulties in accurately and efficiently predicting normals for various point clouds. In this work, we present a new idea of feature extraction for robust normal estimation of point clouds. We use the fusion of multi-scale features from different neighborhood sizes to address the issue of selecting reasonable patch sizes for various data or geometries. We seek to model a patch feature fitting (PFF) based on multi-scale features to approximate the optimal geometric description for normal estimation and implement the approximation process via multi-scale feature aggregation and cross-scale feature compensation. The feature aggregation module progressively aggregates the patch features of different scales to the center of the patch and shrinks the patch size by removing points far from the center. It not only enables the network to precisely capture the structure characteristic in a wide range, but also describes highly detailed geometries. The feature compensation module ensures the reusability of features from earlier layers of large scales and reveals associated information in different patch sizes. Our approximation strategy based on aggregating the features of multiple scales enables the model to achieve scale adaptation of varying local patches and deliver the optimal feature description. Extensive experiments demonstrate that our method achieves state-of-the-art performance on both synthetic and real-world datasets with fewer network parameters and running time.
Read the full abstract

All authors Qing Li, Huifang Feng, Kanle Shi, Yue Gao, Yi Fang, Yu-Shen Liu, Zhizhong Han

Computer Vision · Cross-listed from cs.LG

BanglaMM-Disaster: A Multimodal Transformer-Based Deep Learning Framework for Multiclass Disaster Classification in Bangla

Natural disasters remain a major challenge for Bangladesh, so real-time monitoring and quick response systems are essential. In this study, we present BanglaMM-Disaster, an end-to-end deep learning-based multimodal framework for disaster classification in Bangla, using both textual and visual data from social media. We constructed a new dataset of 5,037 Bangla social media posts, each consisting of a caption and a corresponding image, annotated into one of nine disaster-related categories. The proposed model integrates transformer-based text encoders, including BanglaBERT, mBERT, and XLM-RoBERTa, with CNN backbones such as ResNet50, DenseNet169, and MobileNetV2, to process the two modalities. Using early fusion, the best model achieves 83.76% accuracy. This surpasses the best text-only baseline by 3.84% and the image-only baseline by 16.91%. Our analysis also shows reduced misclassification across all classes, with noticeable improvements for ambiguous examples. This work fills a key gap in Bangla multimodal disaster analysis and demonstrates the benefits of combining multiple data types for real-time disaster response in low-resource settings.
Read the full abstract

All authors Ariful Islam, Md Rifat Hossen, Md. Mahmudul Arif, Abdullah Al Noman, Md Arifur Rahman

Computer Vision

SurgMLLMBench: A Multimodal Large Language Model Benchmark Dataset for Surgical Scene Understanding

Recent advances in multimodal large language models (LLMs) have highlighted their potential for medical and surgical applications. However, existing surgical datasets predominantly adopt a Visual Question Answering (VQA) format with heterogeneous taxonomies and lack support for pixel-level segmentation, limiting consistent evaluation and applicability. We present SurgMLLMBench, a unified multimodal benchmark explicitly designed for developing and evaluating interactive multimodal LLMs for surgical scene understanding, including the newly collected Micro-surgical Artificial Vascular anastomosIS (MAVIS) dataset. It integrates pixel-level instrument segmentation masks and structured VQA annotations across laparoscopic, robot-assisted, and micro-surgical domains under a unified taxonomy, enabling comprehensive evaluation beyond traditional VQA tasks and richer visual-conversational interactions. Extensive baseline experiments show that a single model trained on SurgMLLMBench achieves consistent performance across domains and generalizes effectively to unseen datasets. SurgMLLMBench will be publicly released as a robust resource to advance multimodal surgical AI research, supporting reproducible evaluation and development of interactive surgical reasoning models.
Read the full abstract

All authors Tae-Min Choi, Tae Kyeong Jeong, Garam Kim, Jaemin Lee, Yeongyoon Koh, In Cheul Choi, Jae-Ho Chung, Jong Woong Park, Juyoun Park

Computer Vision

Hybrid SIFT-SNN for Efficient Anomaly Detection of Traffic Flow-Control Infrastructure

This paper presents the SIFT-SNN framework, a low-latency neuromorphic signal-processing pipeline for real-time detection of structural anomalies in transport infrastructure. The proposed approach integrates Scale-Invariant Feature Transform (SIFT) for spatial feature encoding with a latency-driven spike conversion layer and a Leaky Integrate-and-Fire (LIF) Spiking Neural Network (SNN) for classification. The Auckland Harbour Bridge dataset is recorded under various weather and lighting conditions, comprising 6,000 labelled frames that include both real and synthetically augmented unsafe cases. The presented system achieves a classification accuracy of 92.3% (+- 0.8%) with a per-frame inference time of 9.5 ms. Achieved sub-10 millisecond latency, combined with sparse spike activity (8.1%), enables real-time, low-power edge deployment. Unlike conventional CNN-based approaches, the hybrid SIFT-SNN pipeline explicitly preserves spatial feature grounding, enhances interpretability, supports transparent decision-making, and operates efficiently on embedded hardware. Although synthetic augmentation improved robustness, generalisation to unseen field conditions remains to be validated. The SIFT-SNN framework is validated through a working prototype deployed on a consumer-grade system and framed as a generalisable case study in structural safety monitoring for movable concrete barriers, which, as a traffic flow-control infrastructure, is deployed in over 20 cities worldwide.
Read the full abstract

Computer Vision

The More, the Merrier: Contrastive Fusion for Higher-Order Multimodal Alignment

Learning joint representations across multiple modalities remains a central challenge in multimodal machine learning. Prevailing approaches predominantly operate in pairwise settings, aligning two modalities at a time. While some recent methods aim to capture higher-order interactions among multiple modalities, they often overlook or insufficiently preserve pairwise relationships, limiting their effectiveness on single-modality tasks. In this work, we introduce Contrastive Fusion (ConFu), a framework that jointly embeds both individual modalities and their fused combinations into a unified representation space, where modalities and their fused counterparts are aligned. ConFu extends traditional pairwise contrastive objectives with an additional fused-modality contrastive term, encouraging the joint embedding of modality pairs with a third modality. This formulation enables ConFu to capture higher-order dependencies, such as XOR-like relationships, that cannot be recovered through pairwise alignment alone, while still maintaining strong pairwise correspondence. We evaluate ConFu on synthetic and real-world multimodal benchmarks, assessing its ability to exploit cross-modal complementarity, capture higher-order dependencies, and scale with increasing multimodal complexity. Across these settings, ConFu demonstrates competitive performance on retrieval and classification tasks, while supporting unified one-to-one and two-to-one retrieval within a single contrastive framework. We release our code and dataset at https://github.com/estafons/confu.
Read the full abstract

All authors Stefanos Koutoupis, Michaela Areti Zervou, Konstantinos Kontras, Maarten De Vos, Panagiotis Tsakalides, Grigorios Tsagkatakis

Computer Vision

HTTM: Head-wise Temporal Token Merging for Faster VGGT

The Visual Geometry Grounded Transformer (VGGT) marks a significant leap forward in 3D scene reconstruction, as it is the first model that directly infers all key 3D attributes (camera poses, depths, and dense geometry) jointly in one pass. However, this joint inference mechanism requires global attention layers that perform all-to-all attention computation on tokens from all views. For reconstruction of large scenes with long-sequence inputs, this causes a significant latency bottleneck. In this paper, we propose head-wise temporal merging (HTTM), a training-free 3D token merging method for accelerating VGGT. Existing merging techniques merge tokens uniformly across different attention heads, resulting in identical tokens in the layers' output, which hinders the model's representational ability. HTTM tackles this problem by merging tokens in multi-head granularity, which preserves the uniqueness of feature tokens after head concatenation. Additionally, this enables HTTM to leverage the spatial locality and temporal correspondence observed at the head level to achieve higher merging ratios with lower merging costs compared to existing methods. Thus, HTTM achieves up to $7\times$ acceleration over the original VGGT with negligible performance drops in a GPU-based inference.
Read the full abstract

All authors Weitian Wang, Lukas Meiner, Rai Shubham, Cecilia De La Parra, Akash Kumar

Computer Vision

CaliTex: Geometry-Calibrated Attention for View-Coherent 3D Texture Generation

Despite major advances brought by diffusion-based models, current 3D texture generation systems remain hindered by cross-view inconsistency – textures that appear convincing from one viewpoint often fail to align across others. We find that this issue arises from attention ambiguity, where unstructured full attention is applied indiscriminately across tokens and modalities, causing geometric confusion and unstable appearance-structure coupling. To address this, we introduce CaliTex, a framework of geometry-calibrated attention that explicitly aligns attention with 3D structure. It introduces two modules: Part-Aligned Attention that enforces spatial alignment across semantically matched parts, and Condition-Routed Attention which routes appearance information through geometry-conditioned pathways to maintain spatial fidelity. Coupled with a two-stage diffusion transformer, CaliTex makes geometric coherence an inherent behavior of the network rather than a byproduct of optimization. Empirically, CaliTex produces seamless and view-consistent textures and outperforms both open-source and commercial baselines.
Read the full abstract

All authors Chenyu Liu, Hongze Chen, Jingzhi Bao, Lingting Zhu, Runze Zhang, Weikai Chen, Zeyu Hu, Yingda Yin, Keyang Luo, Xin Wang

Computer Vision

PathMamba: A Hybrid Mamba-Transformer for Topologically Coherent Road Segmentation in Satellite Imagery

Achieving both high accuracy and topological continuity in road segmentation from satellite imagery is a critical goal for applications ranging from urban planning to disaster response. State-of-the-art methods often rely on Vision Transformers, which excel at capturing global context, yet their quadratic complexity is a significant barrier to efficient deployment, particularly for on-board processing in resource-constrained platforms. In contrast, emerging State Space Models like Mamba offer linear-time efficiency and are inherently suited to modeling long, continuous structures. We posit that these architectures have complementary strengths. To this end, we introduce PathMamba, a novel hybrid architecture that integrates Mamba's sequential modeling with the Transformer's global reasoning. Our design strategically uses Mamba blocks to trace the continuous nature of road networks, preserving topological structure, while integrating Transformer blocks to refine features with global context. This approach yields topologically superior segmentation maps without the prohibitive scaling costs of pure attention-based models. Our experiments on the DeepGlobe Road Extraction and Massachusetts Roads datasets demonstrate that PathMamba sets a new state-of-the-art. Notably, it significantly improves topological continuity, as measured by the APLS metric, setting a new benchmark while remaining computationally competitive.
Read the full abstract

Computer Vision

Co-Training Vision Language Models for Remote Sensing Multi-task Learning

With Transformers achieving outstanding performance on individual remote sensing (RS) tasks, we are now approaching the realization of a unified model that excels across multiple tasks through multi-task learning (MTL). Compared to single-task approaches, MTL methods offer improved generalization, enhanced scalability, and greater practical applicability. Recently, vision language models (VLMs) have achieved promising results in RS image understanding, grounding, and ultra-high-resolution (UHR) image reasoning, respectively. Moreover, the unified text-based interface demonstrates significant potential for MTL. Hence, in this work, we present RSCoVLM, a simple yet flexible VLM baseline for RS MTL. Firstly, we create the data curation engine, including data acquisition, offline processing and integrating, as well as online loading and weighting. This data engine effectively addresses complex RS data enviroment and generates flexible vision-language conversations. Furthermore, we propose a unified dynamic-resolution strategy to address the diverse image scales inherent in RS imagery. For UHR images, we introduce the Zoom-in Chain mechanism together with its corresponding dataset, LRS-VQA-Zoom. The strategies are flexible and effectively mitigate the computational burdens. Additionally, we significantly enhance the model's object detection capability and propose a novel evaluation protocol that ensures fair comparison between VLMs and conventional detection models. Extensive experiments demonstrate that RSCoVLM achieves state-of-the-art performance across diverse tasks, outperforming existing RS VLMs and even rivaling specialized expert models. All the training and evaluating tools, model weights, and datasets have been fully open-sourced to support reproducibility. We expect that this baseline will promote further progress toward general-purpose RS models.
Read the full abstract

All authors Qingyun Li, Shuran Ma, Junwei Luo, Yi Yu, Yue Zhou, Fengxiang Wang, Xudong Lu, Xiaoxing Wang, Xin He, Yushi Chen, Xue Yang

Computer Vision · Cross-listed from cs.SD

Multi-Reward GRPO for Stable and Prosodic Single-Codebook TTS LLMs at Scale

Recent advances in Large Language Models (LLMs) have transformed text-to-speech (TTS) synthesis, inspiring autoregressive frameworks that represent speech as sequences of discrete codec tokens. Among them, single-codebook TTS LLMs have emerged as compact and streamable architectures that jointly model semantic and acoustic integration. However, despite their efficiency, these models often exhibit unstable prosody, speaker drift, and degraded naturalness. To address these issues, we propose a multi-reward Group Relative Policy Optimization (GRPO) framework that directly optimizes the token generation policy of single-codebook TTS LLMs. Beyond standard intelligibility and speaker similarity objectives, our design integrates three rule-based rewards: a length penalty for duration consistency, an entropy regularization reward for decoding stability, and an LLM-annotated prosody alignment reward that explicitly supervises rhythm. In this prosody reward, an external reasoning LLM predicts multiple plausible pause structures via in-context learning, providing a human-preference-aligned supervisory signal for GRPO training. To assess universality, we further attach a flow-matching (FM) decoder on top of the GRPO-optimized AR backbone and observe consistent additional gains, indicating that our reinforcement optimization enhances the intrinsic AR policy. We further conduct a scalability analysis across data sizes and model scales, revealing that the proposed method consistently enhances prosodic stability, speaker similarity, and overall speech naturalness in single-codebook TTS LLMs.
Read the full abstract

Computer Vision

Unlocking Zero-shot Potential of Semi-dense Image Matching via Gaussian Splatting

Learning-based image matching critically depends on large-scale, diverse, and geometrically accurate training data. 3D Gaussian Splatting (3DGS) enables photorealistic novel-view synthesis and thus is attractive for data generation. However, its geometric inaccuracies and biased depth rendering currently prevent robust correspondence labeling. To address this, we introduce MatchGS, the first framework designed to systematically correct and leverage 3DGS for robust, zero-shot image matching. Our approach is twofold: (1) a geometrically-faithful data generation pipeline that refines 3DGS geometry to produce highly precise correspondence labels, enabling the synthesis of a vast and diverse range of viewpoints without compromising rendering fidelity; and (2) a 2D-3D representation alignment strategy that infuses 3DGS' explicit 3D knowledge into the 2D matcher, guiding 2D semi-dense matchers to learn viewpoint-invariant 3D representations. Our generated ground-truth correspondences reduce the epipolar error by up to 40 times compared to existing datasets, enable supervision under extreme viewpoint changes, and provide self-supervisory signals through Gaussian attributes. Consequently, state-of-the-art matchers trained solely on our data achieve significant zero-shot performance gains on public benchmarks, with improvements of up to 17.7%. Our work demonstrates that with proper geometric refinement, 3DGS can serve as a scalable, high-fidelity, and structurally-rich data source, paving the way for a new generation of robust zero-shot image matchers.
Read the full abstract

Computer Vision

LaGen: Towards Autoregressive LiDAR Scene Generation

Generative world models for autonomous driving (AD) are of great value in applications such as data augmentation, closed-loop simulation, and safety-critical scenario evaluation. Unlike the widely studied image modality, in this work we explore generative world models for LiDAR data. Existing generation methods for LiDAR predominantly focus on single frame generation or lack the capacity for interactive simulation, while existing prediction approaches require multiple frames of historical input and can only deterministically predict multiple frames at once. Both paradigms fail to support long-horizon interactive generation. To this end, we introduce LaGen, which, to the best of our knowledge is the first autoregressive framework capable of generating long-horizon LiDAR scenes in a frame-by-frame, interactive manner. LaGen is able to take a single-frame input as a starting point and effectively utilize bounding box information as conditions to generate high-fidelity 4D scene. In addition, we introduce a scene decoupling estimation module to enhance the model's interactive generation capability for object-level content, as well as a noise modulation module to mitigate error accumulation during long-horizon generation. We extensively evaluate LaGen's performance in controlled data generation and long-horizon scene generation on the nuScenes dataset. The experimental results demonstrate that LaGen achieves state-of-the-art performance, especially on later frames. The code is publicly available at: https://github.com/szzhou88/LaGen.
Read the full abstract

All authors Sizhuo Zhou, Xiaosong Jia, Fanrui Zhang, Junjie Li, Juyong Zhang, Yukang Feng, Jianwen Sun, Songbur Wong, Junqi You, Junchi Yan

Computer Vision

AVFakeBench: A Comprehensive Audio-Video Forgery Detection Benchmark for AV-LMMs

The threat of Audio-Video (AV) forgery is rapidly evolving beyond human-centric deepfakes to include more diverse manipulations across complex natural scenes. However, existing benchmarks are still confined to DeepFake-based forgeries and single-granularity annotations, thus failing to capture the diversity and complexity of real-world forgery scenarios. To address this, we introduce AVFakeBench, the first comprehensive audio-video forgery detection benchmark that spans rich forgery semantics across both human subject and general subject. AVFakeBench comprises 12K carefully curated audio-video questions, covering seven forgery types and four levels of annotations. To ensure high-quality and diverse forgeries, we propose a multi-stage hybrid forgery framework that integrates proprietary models for task planning with expert generative models for precise manipulation. The benchmark establishes a multi-task evaluation framework covering binary judgment, forgery types classification, forgery detail selection, and explanatory reasoning. We evaluate 11 Audio-Video Large Language Models (AV-LMMs) and 2 prevalent detection methods on AVFakeBench, demonstrating the potential of AV-LMMs as emerging forgery detectors while revealing their notable weaknesses in fine-grained perception and reasoning.
Read the full abstract

All authors Shuhan Xia, Peipei Li, Xuannan Liu, Dongsen Zhang, Xinyu Guo, Zekun Li

Computer Vision

Shift-Equivariant Complex-Valued Convolutional Neural Networks

Convolutional neural networks have shown remarkable performance in recent years on various computer vision problems. However, the traditional convolutional neural network architecture lacks a critical property: shift equivariance and invariance, broken by downsampling and upsampling operations. Although data augmentation techniques can help the model learn the latter property empirically, a consistent and systematic way to achieve this goal is by designing downsampling and upsampling layers that theoretically guarantee these properties by construction. Adaptive Polyphase Sampling (APS) introduced the cornerstone for shift invariance, later extended to shift equivariance with Learnable Polyphase up/downsampling (LPS) applied to real-valued neural networks. In this paper, we extend the work on LPS to complex-valued neural networks both from a theoretical perspective and with a novel building block of a projection layer from $\mathbb{C}$ to $\mathbb{R}$ before the Gumbel Softmax. We finally evaluate this extension on several computer vision problems, specifically for either the invariance property in classification tasks or the equivariance property in both reconstruction and semantic segmentation problems, using polarimetric Synthetic Aperture Radar images.
Read the full abstract

All authors Quentin Gabot, Teck-Yian Lim, Jérémy Fix, Joana Frontera-Pons, Chengfang Ren, Jean-Philippe Ovarlez

Computer Vision

FIELDS: Face reconstruction with accurate Inference of Expression using Learning with Direct Supervision

Monocular 3D face reconstruction estimates the parameters of a 3D morphable face model (3DMM) from a single image, providing an explicit geometric representation for facial expression analysis. Owing to the scarcity of paired 2D images and 3D facial meshes, most methods are trained in a self-supervised manner by rendering the predicted face and minimizing image-level reconstruction losses, such as photometric and landmark errors, against the input image. However, these losses enforce visual alignment but do not ensure that the predicted parameters preserve affective information. Recent methods add image-level emotion consistency loss, but it can be satisfied by exaggerating facial deformations. To address these issues, we propose FIELDS (Face reconstruction with accurate Inference of Expression using Learning with Direct Supervision), which learns affect-informative expression representations in the FLAME parameter space. Specifically, FIELDS augments 2D self-supervised reconstruction with scan-derived expression anchoring and a parameter-space emotion loss. During training, scan-fitted targets directly constrain expression-related FLAME parameters to discourage exaggerated deformations, while a jointly trained emotion head uses discrete emotion and continuous valence–arousal annotations to make the predicted parameters informative for affect analysis. Experiments show that FIELDS improves affective utility from the predicted parameters in both within- and cross-dataset settings. Notably, it also achieves lower reconstruction errors overall and across most facial regions than the comparison baselines under both rigid and similarity alignment.
Read the full abstract

Computer Vision

3-Tracer: A Tri-level Temporal-Aware Framework for Audio Forgery Detection and Localization

Recently, partial audio forgery has emerged as a new form of audio manipulation. Attackers selectively modify partial but semantically critical frames while preserving the overall perceptual authenticity, making such forgeries particularly difficult to detect. Existing methods focus on independently detecting whether a single frame is forged, lacking the hierarchical structure to capture both transient and sustained anomalies across different temporal levels. To address these limitations, We identify three key levels relevant to partial audio forgery detection and present T3-Tracer, the first framework that jointly analyzes audio at the frame, segment, and audio levels to comprehensively detect forgery traces. T3-Tracer consists of two complementary core modules: the Frame-Audio Feature Aggregation Module (FA-FAM) and the Segment-level Multi-Scale Discrepancy-Aware Module (SMDAM). FA-FAM is designed to detect the authenticity of each audio frame. It combines both frame-level and audio-level temporal information to detect intra-frame forgery cues and global semantic inconsistencies. To further refine and correct frame detection, we introduce SMDAM to detect forgery boundaries at the segment level. It adopts a dual-branch architecture that jointly models frame features and inter-frame differences across multi-scale temporal windows, effectively identifying abrupt anomalies that appeared on the forged boundaries. Extensive experiments conducted on three challenging datasets demonstrate that our approach achieves state-of-the-art performance.
Read the full abstract

All authors Shuhan Xia, Xuannan Liu, Xing Cui, Peipei Li

Computer Vision

From Diffusion to One-Step Generation: A Comparative Study of Flow-Based Models with Application to Image Inpainting

We present a comprehensive comparative study of three generative modeling paradigms: Denoising Diffusion Probabilistic Models (DDPM), Conditional Flow Matching (CFM), and MeanFlow. While DDPM and CFM require iterative sampling, MeanFlow enables direct one-step generation by modeling the average velocity over time intervals. We implement all three methods using a unified TinyUNet architecture (<1.5M parameters) on CIFAR-10, demonstrating that CFM achieves an FID of 24.15 with 50 steps, significantly outperforming DDPM (FID 402.98). MeanFlow achieves FID 29.15 with single-step sampling – a 50X reduction in inference time. We further extend CFM to image inpainting, implementing mask-guided sampling with four mask types (center, random bbox, irregular, half). Our fine-tuned inpainting model achieves substantial improvements: PSNR increases from 4.95 to 8.57 dB on center masks (+73%), and SSIM improves from 0.289 to 0.418 (+45%), demonstrating the effectiveness of inpainting-aware training.
Read the full abstract

Computer Vision

Towards an Effective Action-Region Tracking Framework for Fine-grained Video Action Recognition

Fine-grained action recognition (FGAR) aims to identify subtle and distinctive differences among fine-grained action categories. However, current recognition methods often capture coarse-grained motion patterns but struggle to identify subtle details in local regions evolving over time. In this work, we introduce the Action-Region Tracking (ART) framework, a novel solution leveraging a query-response mechanism to discover and track the dynamics of distinctive local details, enabling effective distinction of similar actions. Specifically, we propose a region-specific semantic activation module that employs discriminative and text-constrained semantics as queries to capture the most action-related region responses in each video frame, facilitating interaction among spatial and temporal dimensions with corresponding video features. The captured region responses are organized into action tracklets, which characterize region-based action dynamics by linking related responses across video frames in a coherent sequence. The text-constrained queries encode nuanced semantic representations derived from textual descriptions of action labels extracted by language branches within Visual Language Models (VLMs). To optimize the action tracklets, we design a multi-level tracklet contrastive constraint among region responses at spatial and temporal levels, enabling effective discrimination within each frame and correlation between adjacent frames. Additionally, a task-specific fine-tuning mechanism refines textual semantics such that semantic representations encoded by VLMs are preserved while optimized for task preferences. Comprehensive experiments on widely used action recognition benchmarks demonstrate the superiority to previous state-of-the-art baselines.
Read the full abstract

All authors Baoli Sun, Yihan Wang, Xinzhu Ma, Zhihui Wang, Kun Lu, Zhiyong Wang

Computer Vision

BotaCLIP: Contrastive Learning for Botany-Aware Representation of Earth Observation Data

Foundation models have demonstrated a remarkable ability to learn rich, transferable representations across diverse modalities such as images, text, and audio. In modern machine learning pipelines, these representations often replace raw data as the primary input for downstream tasks. In this paper, we address the challenge of adapting a pre-trained foundation model to inject domain-specific knowledge, without retraining from scratch or incurring significant computational costs. To this end, we introduce BotaCLIP, a lightweight multimodal contrastive framework that adapts a pre-trained Earth Observation foundation model (DOFA) by aligning high-resolution aerial imagery with botanical relevés. Unlike generic embeddings, BotaCLIP internalizes ecological structure through contrastive learning with a regularization strategy that mitigates catastrophic forgetting. Once trained, the resulting embeddings serve as transferable representations for downstream predictors. Motivated by real-world applications in biodiversity modeling, we evaluated BotaCLIP representations in three ecological tasks: plant presence prediction, butterfly occurrence modeling, and soil trophic group abundance estimation. The results showed consistent improvements over those derived from DOFA and supervised baselines. More broadly, this work illustrates how domain-aware adaptation of foundation models can inject expert knowledge into data-scarce settings, enabling frugal representation learning.
Read the full abstract

All authors Selene Cerna, Sara Si-Moussi, Wilfried Thuiller, Hadrien Hendrikx, Vincent Miele

Computer Vision

You Can Trust Your Clustering Model: A Parameter-free Self-Boosting Plug-in for Deep Clustering

Recent deep clustering models have produced impressive clustering performance. However, a common issue with existing methods is the disparity between global and local feature structures. While local structures typically show strong consistency and compactness within class samples, global features often present intertwined boundaries and poorly separated clusters. Motivated by this observation, we propose DCBoost, a parameter-free plug-in designed to enhance the global feature structures of current deep clustering models. By harnessing reliable local structural cues, our method aims to elevate clustering performance effectively. Specifically, we first identify high-confidence samples through adaptive $k$-nearest neighbors-based consistency filtering, aiming to select a sufficient number of samples with high label reliability to serve as trustworthy anchors for self-supervision. Subsequently, these samples are utilized to compute a discriminative loss, which promotes both intra-class compactness and inter-class separability, to guide network optimization. Extensive experiments across various benchmark datasets showcase that our DCBoost significantly improves the clustering performance of diverse existing deep clustering models. Notably, our method improves the performance of current state-of-the-art baselines (e.g., ProPos) by more than 3% and amplifies the silhouette coefficient by over $7\times$. Code is available at <https://github.com/l-h-y168/DCBoost>.
Read the full abstract

All authors Hanyang Li, Yuheng Jia, Hui Liu, Junhui Hou

Computer Vision

When Robots Obey the Patch: Universal Transferable Patch Attacks on Vision-Language-Action Models

Vision-Language-Action (VLA) models are vulnerable to adversarial attacks, yet universal and transferable attacks remain underexplored, as most existing patches overfit to a single model and fail in black-box settings. To address this gap, we present a systematic study of universal, transferable adversarial patches against VLA-driven robots under unknown architectures, finetuned variants, and sim-to-real shifts. We introduce UPA-RFAS (Universal Patch Attack via Robust Feature, Attention, and Semantics), a unified framework that learns a single physical patch in a shared feature space while promoting cross-model transfer. UPA-RFAS combines (i) a feature-space objective with an $\ell_1$ deviation prior and repulsive InfoNCE loss to induce transferable representation shifts, (ii) a robustness-augmented two-phase min-max procedure where an inner loop learns invisible sample-wise perturbations and an outer loop optimizes the universal patch against this hardened neighborhood, and (iii) two VLA-specific losses: Patch Attention Dominance to hijack text$\to$vision attention and Patch Semantic Misalignment to induce image-text mismatch without labels. Experiments across diverse VLA models, manipulation suites, and physical executions show that UPA-RFAS consistently transfers across models, tasks, and viewpoints, exposing a practical patch-based attack surface and establishing a strong baseline for future defenses.
Read the full abstract

All authors Hui Lu, Yi Yu, Yiming Yang, Chenyu Yi, Qixin Zhang, Bingquan Shen, Alex C. Kot, Xudong Jiang

Computer Vision

Scenes as Tokens: Multi-Scale Normal Distributions Transform Tokenizer for General 3D Vision-Language Understanding

Recent advances in 3D vision-language models (VLMs) highlight a strong potential for 3D scene understanding and reasoning. However, effectively tokenizing 3D scenes into holistic scene tokens, and leveraging these tokens across diverse 3D understanding tasks, remain highly challenging. We present NDTokenizer3D, a generalist 3D VLM that performs a wide range of 3D scene understanding tasks while naturally supporting human interactions, thereby bridging language-level reasoning with 3D spatial understanding. The core of our approach is a novel three-stage scene tokenization pipeline built upon a Multi-Scale Normal Distributions Transform (NDT) representation, paired with a Multi-Scale NDT Decoder (MSDec). Specifically, NDTokenizer3D first constructs a multi-scale NDT representation from raw high-resolution point clouds, preserving both global context and fine-grained geometric details. Next, the MSDec progressively fuses cross-scale NDT features, producing holistic scene tokens consumable by LLM endpoints. Beyond tokenization, MSDec is repurposed as a general interface for human-interactive prompting (points, boxes, masks) and segmentation-mask decoding, unifying diverse 3D scene understanding tasks within a single architecture. With this compact and unified design, NDTokenizer3D offers a fine-grained, general-purpose 3D VLM, achieving remarkable improvements in 3D Referring Segmentation, 3D Visual Question Answering, and 3D Dense Captioning.
Read the full abstract

All authors Yutao Tang, Cheng Zhao, Gaurav Mittal, Rohith Kukkala, Rama Chellappa, Cheng Peng, Mei Chen

Computer Vision

AnchorOPT: Towards Optimizing Dynamic Anchors for Adaptive Prompt Learning

Existing prompt learning methods, which are built upon CLIP models, leverage textual tokens as anchors to guide the learnable soft tokens. This guidance improves CLIP generalizations. However, these anchors-static in both value and position-lack cross-task and stage-adaptive flexibility. To address this limitation, we propose AnchorOPT, a dynamic anchor-based prompt learning framework. Specifically, AnchorOPT introduces dynamism in two key dimensions: (i) anchor values eschew handcrafted explicit textual tokens (e.g., "shape", "color"), instead learning dynamically from task-specific data; and (ii) the positional relationship between anchor and soft tokens is no longer fixed but adaptively optimized via a learnable position matrix conditioned on the training stage and task context. Training occurs in two stages: we first learn the anchor tokens, then freeze and transfer them to the second stage for optimization of soft tokens and the position matrix. Extensive experiments demonstrate that using only a simple learnable anchor and position matrix achieves performance comparable to or exceeding some methods incorporating additional learnable modules or regularization techniques. As a plug-and-play module, AnchorOPT integrates seamlessly into existing frameworks, yielding consistent performance gains across diverse datasets. Code is publicly available at https://github.com/zhengli97/ATPrompt.
Read the full abstract

All authors Zheng Li, Yibing Song, Xin Zhang, Lei Luo, Xiang Li, Jian Yang

Computer Vision

Progress by Pieces: Test-Time Scaling for Autoregressive Image Generation

Recent visual autoregressive (AR) models have shown promising capabilities in text-to-image generation, operating in a manner similar to large language models. While test-time computation scaling has brought remarkable success in enabling reasoning-enhanced outputs for challenging natural language tasks, its adaptation to visual AR models remains unexplored and poses unique challenges. Naively applying test-time scaling strategies such as Best-of-N can be suboptimal: they consume full-length computation on erroneous generation trajectories, while the raster-scan decoding scheme lacks a blueprint of the entire canvas, limiting scaling benefits as only a few prompt-aligned candidates are generated. To address these, we introduce GridAR, a test-time scaling framework designed to elicit the best possible results from visual AR models. GridAR employs a grid-partitioned progressive generation scheme in which multiple partial candidates for the same position are generated within a canvas, infeasible ones are pruned early, and viable ones are fixed as anchors to guide subsequent decoding. Coupled with this, we present a layout-specified prompt reformulation strategy that inspects partial views to infer a feasible layout for satisfying the prompt. The reformulated prompt then guides subsequent image generation to mitigate the blueprint deficiency. Together, GridAR achieves higher-quality results under limited test-time scaling: with N=4, it even outperforms Best-of-N (N=8) by 14.4% on T2I-CompBench++ while reducing cost by 25.6%. It also generalizes to autoregressive image editing, showing comparable edit quality and a 13.9% gain in semantic preservation on PIE-Bench over larger-N baselines.
Read the full abstract

All authors Joonhyung Park, Hyeongwon Jang, Joowon Kim, Eunho Yang

Computer Vision

LLaVA-UHD v3: Progressive Visual Compression for Efficient Native-Resolution Encoding in MLLMs

Visual encoding followed by token condensing has become the standard architectural paradigm in multi-modal large language models (MLLMs). Many recent MLLMs increasingly favor global native- resolution visual encoding over slice-based methods. To investigate this trend, we systematically compare their behavior on vision-language understanding and attention patterns, revealing that global encoding enhances overall capability but at the expense of greater computational overhead. To address this issue, we present LLaVA-UHD v3, an MLLM centered upon our proposed Progressive Visual Compression (PVC) method, which can be seamlessly integrated into standard Vision Transformer (ViT) to enable efficient native-resolution encoding. The PVC approach consists of two key modules: (i) refined patch embedding, which supports flexible patch-size scaling for fine-grained visual model- ing, (ii) windowed token compression, hierarchically deployed across ViT layers to progressively aggregate local token representations. Jointly modulated by these two modules, a widely pretrained ViT can be reconfigured into an efficient architecture while largely preserving generality. Evaluated across extensive benchmarks, the transformed ViT, termed ViT-UHD, demonstrates competitive performance with MoonViT while reducing TTFT (time-to-first-token) by 2.4x, when developed within an identical MLLM architecture. Building upon ViT-UHD, LLaVA-UHD v3 also achieves competitive performance to Qwen2-VL, while further reducing TTFT by 1.9x. We will release all code and checkpoints to support future research on efficient MLLMs.
Read the full abstract

All authors Shichu Sun, Yichen Zhang, Haolin Song, Zonghao Guo, Chi Chen, Yidan Zhang, Yuan Yao, Zhiyuan Liu, Maosong Sun