Peer Revenue Publish or profit

Papers

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

Computer Science Multimedia

cs.MM Watch Multimedia

Wednesday 26 August

Multimedia · Cross-listed from cs.CR

MMJailBench: A Factorized Benchmark for Disentangling Multimodal Jailbreak Vulnerabilities

Tianshi Wang, Jingsong Wang, Yafei Huang et al. (+3 more)

Multimodal Large Language Models (MLLMs) are increasingly deployed in real-world applications, yet how different factors shape their jailbreak vulnerabilities remains poorly understood. Existing benchmarks often couple harmful intent, prompt framing, visual semantics, and instruction carrier within individual jailbreak instances, obscuring the specific sources of observed vulnerabilities. To address this limitation, we introduce MMJailBench, a factorized benchmark that systematically varies and combines these factors under controlled configurations, enabling fine-grained comparison and factor-level attribution. Large-scale evaluations across 16 open-weight and proprietary MLLMs reveal highly heterogeneous and model-dependent vulnerability profiles. Jailbreak vulnerability varies markedly across harm domains, exposing uneven coverage in current multimodal safety alignment. Prompt framing emerges as the dominant source of variation, task-relevant visual semantics systematically increase jailbreak susceptibility with authority-like cues exposing particularly pronounced vulnerabilities, and visually rendered instructions do not consistently increase jailbreak susceptibility relative to direct textual instructions. To further investigate the risks introduced by multimodal context, we conduct diagnostic analyses on a representative open-weight model and identify vulnerability-associated patterns in internal representations and cross-modal interactions. Finally, we develop a modular multimodal jailbreak evaluation suite with full and lightweight configurations, multiple judge options, and multidimensional metrics, enabling reproducible, scalable, and cost-efficient multimodal jailbreak auditing.
Read the full abstract

All authors Tianshi Wang, Jingsong Wang, Yafei Huang, Fengling Li, Xin Li, Lei Zhu

Tuesday 25 August

Multimedia · Cross-listed from cs.CV

See More, Detect Less? Taming Information Leakage in Multi-View Anomaly Detection

Shang-Fu Chen, Kuan-Chuan Peng, Jhih-Ciang Wu et al. (+2 more)

In multi-view anomaly detection, more cross-view information can actually hurt. When multiple inspection views are naively fused in a reconstruction-based pipeline, normal cues from intact views propagate to the decoder, which faithfully reconstructs anomalous regions, collapsing the reconstruction gap the detector depends on. We call this failure mode cross-view information leakage and show that effective multi-view fusion must explicitly restrict the information reaching the decoder. Building on this insight, we present GLAD(Global-Local Attention Driven framework), the first framework combining vision foundation model features with local and global cross-view fusion for multi-view anomaly detection. The Multi-view Merging Attention (MMA) module performs local cross-view fusion at linear complexity with learnable view importance weighting and token-wise gating, letting each view selectively incorporate fine-grained evidence from other views at $\mathcal{O}(N)$ cost. The Object-Guided Attention (OGA) module captures global context by aggregating class tokens from all views into a single object-level representation and broadcasting it back to patch tokens via temperature-scaled sigmoid gating, replacing the original patch representations rather than adding a residual to preserve the reconstruction gap. Experiments on Real-IAD and MANTA-Tiny show that GLAD outperforms state-of-the-art methods across sample-, image-, and pixel-level metrics, confirming that principled information restriction is key to multi-view anomaly reasoning.
Read the full abstract

All authors Shang-Fu Chen, Kuan-Chuan Peng, Jhih-Ciang Wu, Wen-Huang Cheng, Kai-Lung Hua

Multimedia · Cross-listed from cs.CV

What Do Audio-Visual Synchronization Metrics Actually Measure?

Jai Kumar Sharma and Peeyush Tapadiya

Automatic AV-sync metrics are widely used to rank and train audio-visual generators, but they are rarely audited as measurement instruments. We jointly audit AV-Align, ImageBind AV-relevance, JavisScore, and Synchformer/DeSync under a common reliability protocol: controlled-distortion monotonicity, preprocessing sensitivity, rank uncertainty, cross-metric agreement, PEAVS-proxy agreement, and learned fusion. The result is an axis split, not a single winner: Synchformer/DeSync is the strongest temporal-offset tracker ($\tau=0.84$), ImageBind/JavisScore better match the PEAVS human-aligned proxy ($\tau=0.20$) and content-disruption families, and AV-Align is the weakest standalone metric. The metrics mutually disagree (Krippendorff $\alpha=0.066$), and neither linear nor simple $k$-NN fusion improves PEAVS agreement over the best individual metric. We recommend reporting AV-sync as a Reliability Card (metric-family breakdowns with confidence intervals) rather than a single bare synchronization score.
Read the full abstract

2608.25157 · 25 Aug · Accepted at the ECCV 2026 Workshop on Generative AI for Audio-Visual Content Creation (Gen4AVC), poster presentation; non-archival workshop. 7 pages (4-page main text + references + 2-page appendix), 3 figures, 8 tables. Project page: https://jaishrm07.github.io/avsync-reliability-card/

Abstract PDF

Multimedia · Cross-listed from cs.AI

PhysElite: How Far Are LLMs from Solving Olympiad-Level Physics Problems?

Ruoran Xu, Wending Gao, Liyunfeng Chen et al. (+5 more)

Understanding how (multimodal) large language models perform on physics problems requires benchmarks that reflect the difficulty and breadth of expert-level physical reasoning. Existing physics benchmarks remain limited in the following two important ways: (1) short of high-difficulty datasets, and (2) lack of comprehensive coverage of visual forms, knowledge points, and step-by-step solution processes. As a result, model performance on current datasets may not be fully representative of their ability to solve complex physics problems. To address these issues, we present PhysElite, a large-scale bilingual multimodal benchmark for Olympiad-level physics reasoning. PhysElite contains 11,586 Olympiad-tier problems. For each problem, we provide corresponding visual diagrams, step-by-step bilingual Chinese-English solution derivations, and the final answer. We benchmark 18 open-source and closed-source MLLMs, and find that even the strongest model reaches only 33.7% answer accuracy. We additionally conduct step-level process evaluation to diagnose where models fail in the reasoning chain. Our datasets are released at https://huggingface.co/datasets/physelite/PhysElite.
Read the full abstract

All authors Ruoran Xu, Wending Gao, Liyunfeng Chen, Aixin Shi, Haoyu Cheng, Zixiang Fang, Yiqiang Zou, Qiufeng Wang

Multimedia · Cross-listed from cs.CL

Unsupervised Post-Training of Foundation Models: A Survey

Yijie Xu, Qianyi Cai, Huizai Yao et al. (+9 more)

Foundation-model post-training usually relies on human labels, preference data, stronger teachers, or executable verifiers. We study Unsupervised Post-Training (UPT): update-bearing adaptation on unlabeled inputs whose learning signal is derived from same-lineage model artifacts rather than an external oracle. We catalog 80 strict UPT methods and organize them by the object that supplies the update signal: a prediction statistic, a sample relation, a self-generated target, or an internal evaluator. Beyond inventory, we show how the choice of internal signal and task structure determines whether post-training improves the model or recursively amplifies error. An orthogonal Input Visibility $\times$ Update Persistence view maps deployment regimes and defines a unified framework for UPT selection and evaluation.
Read the full abstract

All authors Yijie Xu, Qianyi Cai, Huizai Yao, Yili Wang, Tianfu Wang, Cehao Yang, Xingbo Yao, Zhiyu Guo, Aiwei Liu, Xuming Hu, Weiyu Guo, Hui Xiong

Multimedia

Learning to Prefer Reliably: Error-Augmented Emotion Preference Optimization with Calibrated Fusion

Zilong Huang, Junyi Peng, Junjie Li et al. (+5 more)

Emotion preference learning uses pairwise comparisons between candidate descriptions to align multimodal large language models (MLLMs) with human judgments of open-ended emotion descriptions and to train reward models that capture human emotional preferences. However, conventional pairwise supervision is often sparse, typically providing only a single negative description for each positive description, and therefore offers limited coverage of the diverse ways in which an emotion description can be incorrect. In particular, models may be insufficiently exposed to semantically fluent but emotionally inconsistent descriptions. Beyond this data-level limitation, relying on a single MLLM judge introduces a distinct model-level concern: its judgments can be affected by model-specific biases when interpreting fine-grained or ambiguous multimodal emotional cues. To address these limitations, we propose Error-Augmented Preference Optimization (EAPO), a framework for improving the reliability of MLLM-based emotion preference judgment at both the data and model levels. First, we construct an error-augmented dataset by generating multiple controlled and emotion-aware negative descriptions from each preferred description. We then adapt multiple independent MLLM judges to this richer supervision and aggregate their preference margins using margin-calibrated soft fusion, which maps heterogeneous margins to a common scale before aggregation. Experiments on the MER2026-EmoPrefer Challenge dataset and our error-augmented dataset demonstrate that EAPO improves emotion preference prediction and enhances the robustness of MLLM judges when evaluating fluent descriptions that conflict with the video's multimodal emotional evidence. Our code is available at https://github.com/slash1028/EAPO-EmoPrefer.
Read the full abstract

All authors Zilong Huang, Junyi Peng, Junjie Li, Kai Li, Wenze Ren, Kong Aik Lee, Man-Wai Mak, Tatsuya Kawahara

Multimedia

EVEREST:Endogenous Vision-Language Reinforcement Reasoning Exploration for Urban Socio-Semantic Segmentation

Qixiu Li, Zhongzhi He, Xiang Zhu et al. (+3 more)

Urban socio-semantic segmentation leverages digital and satellite imagery to provide critical spatial semantic information for downstream applications such as urban resource allocation. Although existing methods achieve high segmentation accuracy, they still suffer from inaccurate delineation of target boundaries. The underlying issue is that current models primarily rely on passively aggregated global cross-modal cues, lacking active exploration of the environment. To address this limitation, we propose the EVEREST model, which adopts an egocentric exploration strategy that enables the model to actively investigate boundary cues and perform self-correction. In addition, we formulate discrete natural-language prompts as pseudocode to regularize the execution logic. Reinforcement learning is further employed to implement this irreducible process and elicit the model's structured reasoning capability. Our EVEREST achieves optimal performance on all metrics in the real world urban socio-semantic dataset, demonstrating the superiority of our model. Codes are available at https://github.com/TechCloud-x/EVEREST.
Read the full abstract

All authors Qixiu Li, Zhongzhi He, Xiang Zhu, Xiaoyong Li, Jiarun Lin, Weifeng Xu

Multimedia

Task-disentangled Low-Rank Adaptation for Versatile Audio-visual Multi-modal Learning Tasks within a Unified Framework

Hanyu Xuan, Mengqi Zhang, Junjun Mao et al. (+5 more)

Inspired by human multi-modal perception, Audio-Visual Multi-Modal Learning (AVMML) integrates auditory and visual information to leverage complementary cross-modal cues, enabling more robust and comprehensive scene perception. Existing studies predominantly tackle each AVMML task in isolation, which stands in stark contrast to humans' unified cognitive capacity for handling versatile perception. However, naive joint training across multiple AVMML tasks often suffers from mutual interference, arising from the intricate inter-task relationships. To address this, we propose a unified framework that simultaneously accommodates versatile AVMML tasks. Specifically, benefiting from powerful representation and generalization capabilities of large language models, we design a task-disentangled Low-Rank Adaptation (LoRA) mechanism that enables dynamic integration of both task-specific and task-shared knowledge, thereby facilitating effective multi-task collaboration. The proposed task-disentangled LoRA comprises three components: a task-general low-rank matrix, task-specific modulation matrices, and cross-task collaboration experts, which respectively capture universal audio-visual knowledge, decouple task-specific pattern, and exploit inherent inter-task correlations. By unifying explicit collaboration from both model and task perspectives, our approach not only surpasses existing unified audio-visual models across multiple AVMML tasks, but also outperforms most task-specific models on certain AVMML tasks.
Read the full abstract

All authors Hanyu Xuan, Mengqi Zhang, Junjun Mao, Fei Wang, Kun Li, Guanghui Yue, Zhiliang Wu, Hehe Fan

Multimedia · Cross-listed from cs.CV

Boot-and-Feedback Framework for Generalist-Expert Model Collaboration in Breast Ultrasound Diagnosis

Ming Cheng, Hongyu Sun, Zhaolin Chen et al. (+3 more)

Breast ultrasound (BUS) is widely used for breast cancer diagnosis yet remains operator-dependent. While deep learning shows promise, ensuring diagnostic reliability and interpretability is challenging. Recent Multimodal Large Language Models (MLLMs) often generate spurious descriptions due to limited domain knowledge, which mislead downstream expert models and compromise clinical validity. To address these challenges, we propose the Boot-and-Feedback (BooF) model collaboration framework for synergistic MLLM-expert interaction. Specifically, in the Boot Stage, the MLLM is guided by the BI-RADS lexicon and preliminary benign-malignant vision-expert predictions, enabling it to transfer general reasoning to BUS analysis while avoiding hallucinations. Subsequently, the Feedback Stage integrates these descriptions with visual features via a lightweight Attention-Gated Cross-Modality Fusion Module. This allows the expert to leverage textual feedback while adaptively filtering noise. Extensive experiments on multiple BUS datasets demonstrate that BooF substantially outperforms state-of-the-art methods in terms of diagnostic accuracy and interpretability.
Read the full abstract

All authors Ming Cheng, Hongyu Sun, Zhaolin Chen, Jun Liu, Hossein Rahmani, Qiuhong Ke

Monday 24 August

Multimedia

Mind the Couch! Eliciting MLLM Reasoning in Interior Design via Weak-to-Strong Task Vector Injection

Yuxuan Yang, Jingyao Wang and Luntian Mou

Multimodal Large Language Models (MLLMs) have demonstrated great performance, yet they often suffer from severe modality misalignment when confronted with densely constrained spaces for interior design. Due to the loss of high-frequency local topological details and fine-grained aesthetic shifts during visual encoding, existing MLLMs frequently hallucinate, yielding physical spatial collisions and visual aesthetic dissonance. To address this, we propose Dual-prior Activation Residual Task-vectors Injection mechanism (DART-I) for MLLMs. It shifts the paradigm from lossy text-prompting to direct latent intervention, utilizing weak-to-strong deterministic rules to anchor the causal reasoning of MLLMs for interior design. Specifically, DART-I operates in three steps: it first explicitly extracts continuous spatial distance and color typography features from images using extremely lightweight weak experts; subsequently, it transforms these deterministic priors into directional task vectors via a linear projection network; these vectors are dynamically injected as residual terms into the latent space of the frozen MLLMs, steering MLLMs towards precise reasoning for interior design. Stepping outside the conventional paradigms, our method achieves precise reasoning without fine-tuning the MLLMs, effectively bypassing expensive computational costs and catastrophic forgetting. Extensive experiments on various benchmarks demonstrate the effectiveness and advantages of DART-I.
Read the full abstract

Multimedia

SACHA: Semantic-Aware Compression for 3D Gaussian Head Avatars

Zihan Zhang, Shanzhi Yin, Xinju Wu et al. (+5 more)

Animatable 3D Gaussian head avatars offer high-fidelity and flexible facial rendering, but typically require substantial storage and transmission costs for numerous Gaussian primitives. Existing Gaussian head avatar methods overlook the visual saliency of different head semantic regions for more appropriate Gaussian primitive allocation, as well as the efficient compression of trained head avatar sequences. To tackle this obstacle, we propose SACHA, a dynamic head avatar compression framework that leverages both semantic-aware density control and appearance-motion decomposition to achieve compact representation and high-quality novel-view rendering of head avatar sequences. Specifically, the semantic-aware density control guides the adaptive allocation of Gaussian primitives across different head regions with region-adaptive densification and pruning. In addition, the appearance-motion decomposed compression further reduces the temporal redundancy of the avatar sequence by transmitting only head-prior parameters for avatar movements. Together, these designs enable a compact representation for efficient transmission of dynamic Gaussian head avatars while preserving visual fidelity. Experiments demonstrate that SACHA achieves a superior rate-distortion performance over existing Gaussian head avatar representation and compression methods while maintaining high-quality novel-view and novel-expression rendering.
Read the full abstract

All authors Zihan Zhang, Shanzhi Yin, Xinju Wu, Bolin Chen, Ru-Ling Liao, Jie Chen, Shiqi Wang, Yan Ye

Multimedia · Cross-listed from cs.CV

Cultural Moment Benchmark: Evaluating Video Cultural Reasoning and Grounding in Southeast Asia

Burak Satar, Zhixin Ma, Cheng Yu-Tong et al. (+3 more)

Cultural understanding in video means more than recognizing what is visible; it requires grasping the symbolic and temporal significance of cultural concepts. We decompose this into three abilities: naming what a concept symbolizes, visually recognizing it on video, and locating its sub-events in time. Existing video-cultural benchmarks tend to test what is seen, collapsing these three abilities into a single score that hides the bottleneck. We introduce the Cultural Moment Benchmark (CMB): 306 expert-curated concepts from seven countries in Southeast Asia across five categories. We evaluate each concept through three stages, one per ability. Given a description, Stage 1 (S1) selects from four candidate concept names, Stage 2 (S2) selects from four candidate video moments, and Stage 3 (S3) predicts the start and end times of the moment in a video. To keep each stage focused on a distinct ability, we use three design choices: semantic-similarity distractors (S1, S2), unlabeled video moments (S2), and free-form localization on a different example video (S3). Across six vision-language models, failure modes vary by ability and modality. i) Even the strongest closed-source models score below 30% when all three stages must be correct; ii) The three abilities do not fully cascade: naming a concept correctly helps half the models recognize it on video, but recognizing it has little effect on locating the sub-event in time; iii) Audio is complementary, redundant, or distracting depending on the concept, more often distracting in non-Latin-script countries; removing both audio and subtitles hurts Games and Music the most. Our 14-rater human study shows that even Expert raters score below chance on concepts from a neighboring country, indicating that CMB requires country-specific cultural knowledge. CMB acts as a diagnostic harness, attributing failures to a specific ability or modality.
Read the full abstract

All authors Burak Satar, Zhixin Ma, Cheng Yu-Tong, Huy Hoang Tran, Phuong Anh Nguyen, Chong-Wah Ngo

Multimedia · Cross-listed from cs.CV

Results of the 1st Asynchronous CASTLE Challenge at the Joint Egocentric Vision Workshop in Conjunction with CVPR 2026

Luca Rossetto, Werner Bailer, Cathal Gurrin et al. (+5 more)

This report summarizes the contributions and results of the 1st Asynchronous CASTLE Challenge at the Joint Egocentric Vision Workshop in conjunction with CVPR 2026.
Read the full abstract

All authors Luca Rossetto, Werner Bailer, Cathal Gurrin, Graham Healy, Omar Shahbaz Khan, Stevan Rudinac, Klaus Schöffmann, Allie Tran

Multimedia

Adaptive Hierarchical Representation Alliance for Multimodal Learning

Chunlei Meng, Pengbin Feng, Jacqueline J. Pang et al. (+5 more)

Multimodal models often align language, vision, and audio in a single final-layer latent space, implicitly assuming that task-relevant evidence emerges at the same semantic depth across modalities. Using layer-wise CKA analysis, we observe that this assumption leads to semantic granularity mismatch: textual cues usually require deeper contextual abstraction, whereas visual and acoustic cues often provide discriminative perceptual evidence in shallow or middle layers. This mismatch can flatten fine-grained modality-private cues and reduce reliability under noisy, imbalanced, or missing inputs. To address this, we proposed Adaptive Hierarchical Representation Alliance (AHRA), a hierarchical shared–private expert framework. AHRA factorizes each modality into shared and private streams across semantic levels, regularizes them with shared alignment and private decorrelation, routes shared information through a cross-modal expert, and enhances task-relevant private tokens with modality-specific experts guided by a sparsity-controlled soft-gating mechanism (foreground exam). A hierarchical co-fusion module then performs intra-level expert coordination and inter-level semantic selection. Experiments on six benchmarks across image-text classification, multimodal intent recognition, and trimodal sentiment analysis show that AHRA consistently improves over strong baselines and remains robust under noisy and missing-modality settings.
Read the full abstract

All authors Chunlei Meng, Pengbin Feng, Jacqueline J. Pang, Chih-Ting Liao, Rong Fu, Zhaolu Kang, Zhongxue Gan, Chun Ouyang

Multimedia · Cross-listed from cs.IR

FashionKG-RAG: Knowledge Graph-Enhanced Retrieval-Augmented Generation for Fashion Question Answering

Yujuan Ding, Linyin Luo, Shijie Wang et al. (+5 more)

Fashion is a knowledge-intensive domain in which effective decision-making depends on integrating multiple types of knowledge. Although Large Language Models (LLMs) have transformed many areas, their application in fashion remains limited by hallucinations and weak domain specialization. Knowledge Graph (KG)-based Retrieval-Augmented Generation (RAG) offers a promising way to add structured knowledge to LLMs. However, existing fashion KGs are typically restricted to product-level attributes or item relations, and fail to capture the broader fashion ecosystem. To bridge these gaps, we propose FashionEcoKG, a comprehensive, domain-wide knowledge graph built with expert-level precision and professionalism. It is constructed through a three-stage agentic pipeline that extracts high-fidelity knowledge cores from authoritative textbooks and strengthens structural connectivity through cross-domain augmentation and generative expansion. To leverage this resource, we further develop PG-RAG (Pruning-Grounding RAG), a training-free framework designed to handle the conceptual density and linguistic noise of fashion queries. Specifically, we introduce a Dual-Granularity Path Re-Ranking (DGPR) module of two stages. The Pruning-based Semantic Ranking (PSR) module distills each query into a skeleton form to improve retrieval recall, while the Grounding-based Agentic Ranking (GAR) performs point-wise scrutiny of candidate paths against the original full query to ensure global relevance. Experiments on a curated fashion QA dataset show that PG-RAG effectively leverages FashionEcoKG to improve retrieval and answer accuracy, outperforming both non-RAG and existing KG-RAG baselines.
Read the full abstract

All authors Yujuan Ding, Linyin Luo, Shijie Wang, Xu Yuan, Yunshan Ma, Yi Bin, Wenqi Fan, Qing Li

Sunday 23 August

Multimedia · Cross-listed from eess.IV

Local SVD-Entropy Maps as a Complementary Structural Representation for Full-Reference and No-Reference Image Quality Assessment

Andrei Velichko and Petr Boriskov

We investigate a local spectral-complexity representation for perceptual image quality assessment (IQA) based on Shannon entropy of singular values computed directly from two-dimensional image patches. For each $3\times3$-pixel grayscale patch, SVD is applied directly and the normalized singular-value entropy defines one HSVD-map value. The construction requires neither flattening nor delay embedding, uses no boundary padding, and is invariant to $90^{\circ}$ rotations and mirror reflections at the local-descriptor level. A nested salt-and-pepper experiment on Lena separates absolute similarity to a clean reference from sensitivity to an additional degradation step. HSVD-SSIM responds more strongly to local corruption and retains a larger neighboring-state response at severe noise levels. Validation on all 10,125 distorted KADID-10k images shows that HSVD-SSIM is weaker than conventional SSIM as a standalone full-reference metric (SRCC $0.450$ vs. $0.619$), but complementary when combined with it: grouped cross-validation increases SRCC from $0.618$ to $0.659$, with a bootstrap 95% confidence interval of $[0.036,0.046]$ for the gain. In a no-reference experiment, adding HSVD-derived single-image descriptors improves the best nonlinear model from SRCC $0.528$ to $0.575$ (95% CI $[0.033,0.061]$) and also improves prediction of quality changes between neighboring distortion states. These results support direct local SVD entropy as an interpretable structural channel that complements conventional image-domain similarity and remains informative without a pristine reference.
Read the full abstract

Multimedia · Cross-listed from cs.MA

CaseWeaver: A Multi-Agent Framework for Multimodal Virtual Clinical Case Generation

Jierui Qu, Jiachuan Peng, Lin Li et al. (+2 more)

Clinical diagnosis relies on consistent multimodal data collected from the same patient throughout the disease course, yet such data are difficult to acquire at scale because of collection costs, missing modalities, fragmented systems, and longitudinal follow-ups. Existing synthetic-data approaches largely focus on individual modalities or vision-language dual modalities at report-level generation. Little work has been done to construct synthetic data with consistent patient backgrounds, coherent disease trajectories, and interrelated modality-specific evidence at a complete clinical case level. We introduce CaseWeaver, a multi-agent framework built around a timeline-anchored Latent Clinical Case Graph (LCCG). The LCCG organizes patient context, latent disease states, clinical events, and expected observations in a shared patient-level representation. Modality-agents use scoped observation subgraphs and clinical protocols to generate evidence including clinical records, laboratory results, physiological signals, and medical images. We evaluate clinical inferability using a calibrated AgentClinic protocol and case diversity using Virtual Case Diversity (VCD) score. CaseWeaver outperformed general-model and agentic-workflow baselines on both metrics, producing more diverse and coherent multimodal virtual clinical cases.
Read the full abstract

All authors Jierui Qu, Jiachuan Peng, Lin Li, Kyle Lam, Jianing Qiu

Multimedia · Cross-listed from cs.RO

The Potential of Haptic Foundation Models

Jianquan Wang, Haiwei Dong and Abdulmotaleb El Saddik

Despite the success of foundation models in language and vision, their expansion into embodied AI is bottlenecked by a lack of generalized touch sensing. This limitation is especially relevant to consumer electronics, where smartphones, wearables, VR controllers, home robots, and health monitoring devices require safe and adaptive physical interaction. Constrained by hardware heterogeneity and the necessity of active physical data collection, current haptic models remain rigidly task-specific. To overcome these limitations, this article explores the transformative potential and developmental trajectory of Haptic Foundation Models (HFMs). We detail the paradigm shift required to transition from passive Large Language Models and Vision Language Models into active HFMs across four core dimensions: action coupling, physical dynamical representation space, continuous time-series data granularity, and action-conditioned future state prediction. Furthermore, we synthesize existing large-scale tactile datasets and benchmark UniTouch, AnyTouch, T3, and Sparsh on TacBench for force estimation, slip detection, and relative pose estimation.
Read the full abstract

Multimedia

Motion-Aware Reasoning from Speech to Mask Tracks: Runner-up Solution for the MeViS-Audio Track of the 8th LSVOS Challenge 2026

Jinxing Zhou, Suiyi Zhao, Yanghao Zhou et al. (+1 more)

Speech-guided referring video object segmentation aims to recover the mask tracks of objects specified by a spoken motion description. Here, speech carries a linguistic instruction rather than acoustic evidence from a sounding object, so a solution must connect speech recognition, motion-centric temporal grounding, mask tracking, and explicit no-target handling. We introduce Speech2MaskTrack, our approach for the MeViS-Audio track of the 8th LSVOS Challenge. Speech2MaskTrack transcribes the spoken query and compiles it into structured constraints over category, count, direction, interaction role, and temporal phase. SAM3.1 enumerates multiple instance tracks, which TRACE ranks using complete-trajectory motion and relation evidence. A frozen lexical presence gate may suppress the ranked SAM3.1 base prediction. When the gate predicts that a target is present, an available full-expression-conditioned SaSaSa2VA track replaces the SAM3.1 mask. Only outputs that remain empty enter GPT-assisted recovery, which invokes SaSaSa2VA again under query- and mask-level verification. Speech2MaskTrack achieved second place in the official challenge ranking.
Read the full abstract

All authors Jinxing Zhou, Suiyi Zhao, Yanghao Zhou, Ruohao Guo

Multimedia

ReART: Reference-Guided Retrieval and Refinement for Emotion-Aware Art Generation

Qianqian Tang, Jiayi Gao, Ting Lei et al. (+1 more)

Emotion-aware artistic image generation requires a model to satisfy semantic content, artistic style, and target emotion simultaneously. The key challenge is that artistic captions conflate these axes into underspecified free-form text, making fine-grained visual attributes such as brushwork, composition, and tonal atmosphere difficult to ground concretely. We present ReART, a reference-guided retrieval and refinement framework. Our method decomposes test captions and each image annotation in the EmoArt database into structured visual fields, and performs field-wise retrieval over subject, layout, brush-line, and tone-mood dimensions to retrieve role-specific visual references that supply the perceptual detail text alone cannot convey; these references are used alongside a structured prompt for initial synthesis. For samples where any Attribute Alignment Score (AAS) axis falls below threshold, an AAS-driven refinement loop diagnoses failures, constructs constrained repair plans specifying elements to keep, errors to fix, and operations to avoid, routes references by correction purpose, and performs controlled editing under structural preservation constraints. Our system ranks 2nd in Track 1 of the AffectiveArt 2026 Grand Challenge, achieving a perfect AAS of 1.00 and an overall score of 0.78. Code is available at https://github.com/oceanflowlab/ReART.git.
Read the full abstract

All authors Qianqian Tang, Jiayi Gao, Ting Lei, Yang Liu

Multimedia · Cross-listed from cs.AI

Beyond What Meets the Eye: Unveiling Situational Illusions for Multimodal Large Language Models

Zhiming Yang, Zhuoxi Xiong, Donglin Zhou et al. (+3 more)

Real-world situation appearances can deviate from their underlying physical states, challenging the reliability of multimodal large language models (MLLMs) in practical applications. In this paper, we term this phenomenon situational illusions and investigate: (1) how MLLMs perform under such illusions, and (2) how to mitigate the limitations. We first develop a comprehensive where-what-how taxonomy that characterizes where situational illusions occur, what targets they take, and how they arise. Building on this taxonomy, we introduce MSIBench, a benchmark designed to assess the discrimination, understanding, and reasoning capabilities of MLLMs under situational illusions. Evaluations of 27 model configurations reveal that current MLLMs are highly vulnerable to these illusions and exhibit 6 typical failure modes related to visual observation, grounding, and reasoning. To mitigate the limitations, we build on the core idea of systematically inspecting and reasoning over visual evidence for contextual understanding, developing prompting for closed-source models and supervised fine-tuning for open-source models, respectively. These two simple yet effective methods improve model performances by 20% at most, suggesting a practical path toward more reliable multimodal perception and reasoning in complex real-world environments.
Read the full abstract

All authors Zhiming Yang, Zhuoxi Xiong, Donglin Zhou, Wenjun Wei, Shiyao Cui, Jinqiao Shi

Multimedia · Cross-listed from cs.AI

Query-Driven Multimodal Information Extraction from Long Documents

Yikai Gao, Ding Xia and Xi Yang

In domain-specific multimodal long documents, images and text jointly convey complex knowledge that cannot be fully captured by plain text alone. However, existing paradigms like DocVQA primarily focus on generating textual answers or localizing evidence regions, rather than outputting query-specific textual attribute values and corresponding images. To address this gap, we propose query-driven image-text joint extraction from long documents, requiring models to output query-requested textual attribute values and corresponding image bounding boxes. Based on challenges related to both user intent and document content, we designed a two-level taxonomy that operates at the query and instance levels. Further, we construct ITJoint, the first high-quality, manually annotated benchmark for this new task, comprising 2,455 pages of domain-specific documents with numerous non-decorative images, 316 queries, and 910 answer instances. Finally, we evaluate representative standalone Vision-Language Models from different providers and further design Q2IT, a multi-agent collaborative framework consisting of three progressively collaborating agents for evidence collection, page selection, and target-image localization. Using a joint evaluation approach that assesses both text extraction and image localization, our experiments show that standalone VLMs struggle with this task, while Q2IT significantly improves performance on ITJoint, although a substantial gap remains toward perfect results.
Read the full abstract

Saturday 22 August

Multimedia · Cross-listed from cs.CV

Towards Bitstream-corrupted Harsh Visual Understanding: Through Bitstream Language Modeling as Robust Semantic Priors

Chaoran Huang, Fangcheng Li, Tianyi Liu et al. (+2 more)

Bitstream-corrupted Harsh Visual Understanding (BcHVU) aims to understand harshly degraded videos originally decoded from a severely corrupted bitstream in real-world multimedia communication. The ill-posed nature of BcHVU poses a major challenge for existing vision models, as even subtle bitstream corruption can lead to irreversible pixel distortion and significant semantic loss. To address these challenges in BcHVU, we propose Bitstream Language Modeling as Robust Semantic Priors (BLMSP), a framework for learning and injecting bitstream-native semantic cues. Our proposed BLMSP framework learns to extract bitstream-native semantic cues by bitstream language modeling, and leverages them as priors by injecting into off-the-shelf vision models of BcHVU tasks. Specifically, we present a Video Bitstream Byte Model (VBBM) that integrates byte-level modeling and cross-codec semantic distillation, enabling it to interpret robust semantics from byte sequences in multiple corrupted bitstream formats. The learned bitstream semantics are leveraged as robust priors and fused into BcHVU model backbones for improving the quality of video restoration, captioning, and human pose estimation. To train BLMSP, we construct a large-scale multi-source Corrupted-bitstream Harsh-video Paired (CHP) dataset containing 607k corrupted bitstream segments and 287k paired harsh video clips. Extensive experimental results show that the learned bitstream priors improve video restoration, captioning, and human pose estimation by 2.51 dB in PSNR, 0.20 in CIDEr, and 0.18 in PCK@0.2 on average, respectively. These results demonstrate that corrupted bitstream can serve as robust semantic priors in solving pixel distortion and semantic loss in BcHVU.
Read the full abstract

All authors Chaoran Huang, Fangcheng Li, Tianyi Liu, Wenyang Liu, Kejun Wu

Friday 21 August

Multimedia · Cross-listed from cs.CV

Recognition-Conditioned Reasoning: A Training-Free Multimodal-LLM Pipeline for Fine-Grained Micro-Action Understanding

Fengshun Wang, Jin'ang Han and Zhigang Tu

Micro-actions are subtle, short, low-amplitude body movements, such as a fidgeting hand or a slight head tilt, that humans perform with little conscious intent yet that reliably leak emotional and psychological state. Understanding them goes beyond assigning a label: a model must also describe which body parts move and reason, faithfully, about why a clip warrants a particular fine-grained category. We present the training-free, prompt-only system that won first place in the fine-grained understanding track (MA-Bench) of the MAC~2026 Micro-Action Challenge, where both fine-tuning and ground-truth supervision are disallowed. Built entirely upon frozen multimodal large language models (MLLMs), the system dynamically routes each of the eight sub-tasks to the MLLM empirically best suited for that task: a discriminative MLLM for closed-ended recognition tasks and a generative MLLM for open-ended description and reasoning tasks. This architecture achieves a statistically significant performance advantage on open-ended tasks, attaining an average score of 2.68 (on a five-point scale) compared to 1.44 for the second-best approach.
Read the full abstract

Multimedia

When Images Look Right and Retrieve Wrong: Coverage-Guided Cross-Scale Re-Indexing for Knowledge-Faithful Generative Perception

Guangyuan Dong, Chuang Liu, Haoyu Wang et al. (+10 more)

Multimodal information systems increasingly route generated visual content back through the same vision-language index that informed its production, so the output must remain retrievable by the queries it was meant to serve. When the scene contains entities at vastly different scales, existing language-guided generators condition on a single, globally pooled text embedding and quietly drop scale-specific concepts, breaking concept-query retrieval even when pixel fidelity is high. We formalise this failure as semantic collapse and propose CERES, a closed-loop multimodal indexing framework that builds a three-level semantic pyramid, mines implicit concepts via a co-occurrence-aware router, performs scale-routed cross-attention into a lightweight U-Net generator, and verifies coverage by re-indexing the generated image with the same frozen VLM. A continuously differentiable soft-Jaccard coverage objective returns dense gradients to the 0.39 M-parameter generator under explicit non-degeneracy conditions, and coverage is verified by an independent DINOv2 linear probe trained only on external scene and object labels. On four pansharpening benchmarks across seven settings, CERES delivers the new state of the art with the largest gains where scale variation is most extreme (+4.64% relative Q2n and +9.7 mAP for DOTA detection). It also improves concept-query retrieval Recall@5 by +14.0 points and image-text mean reciprocal rank by 0.19 over the strongest baseline, showing that the closed loop preserves queryable content rather than self-referential feature consistency.
Read the full abstract

All authors Guangyuan Dong, Chuang Liu, Haoyu Wang, Yangchen Zeng, Jiaqi Zhang, Li Jiuxing, Xiaoyang Yu, Pinlong Zhao, Yuchao Hou, Ziwei Li, Zheng Lin, Alexander Lim Han Yang, Yusen Wu

Thursday 20 August

Multimedia

KeyBound: Keyed and Host-Bound Learned Audio Watermarking for Speech Provenance

Bangshuo Zhu, Yuxin Cao, Weifei Jin et al. (+4 more)

Audio watermarking is a proactive route to attributing synthetic speech to its source. Learned audio watermarks are typically judged by payload recovery after a fixed catalog of signal distortions such as noise, compression, filtering, and resampling. That test is necessary but not sufficient for provenance. A mark offered as evidence of origin should not be readable by an unauthorized party, should not be transferable to unrelated audio, and should not vanish when the recording is re-synthesized by a modern generative model. We present KeyBound, a learned audio watermark that restores the two ingredients classical watermarking supplied and learned schemes set aside, a secret key and a host-aware carrier. KeyBound masks the payload with a secret key and embeds the masked bits through a carrier modulated by a frozen spectral representation of the host, so the key governs payload access while the host-conditioned carrier resists direct transplantation. A key-independent presence head lets any party detect a mark, whereas only a key holder reads its attribution, and under the single-sample uniformity assumption a wrong-key decode clears our verification rule with probability at most $2.1\times10^{-3}$. On LibriSpeech against WavMark, AudioSeal, and Timbre, KeyBound holds 1.00 detection accuracy and 0.98 bit accuracy under a spectral denoiser that costs every baseline its detection, decodes at chance without the key, and rejects transplanted carriers. Detection further transfers to held-out DAC and BigVGAN re-synthesis, though exact payload recovery degrades. Speech provenance is thus better posed as a keyed, host-bound attribution problem than as the recovery of a payload under a catalog of signal distortions fixed in advance.
Read the full abstract

All authors Bangshuo Zhu, Yuxin Cao, Weifei Jin, Fusen Guo, Huadong Mo, Jingling Xue, Wei Song

Multimedia · Cross-listed from cs.RO

Humanoid Musical Robots as Experimental Interfaces for Music-Evoked Emotion

Vincent K.M. Cheung and Jia-Yeu Lin

Advances in technology have led to increasingly sophisticated musical humanoid robots. However, their use has largely been limited to performance and related research in human-robot interaction. In this position paper, we propose a novel perspective: musical humanoid robots as experimental interfaces for investigating music-evoked emotions. We argue that current research is constrained by paradigms relying on pre-recorded auditory stimuli, which fail to capture the multimodal, embodied, and interactive nature of real-world musical experience. Building on existing theories of music cognition and emotion, we identify mechanisms that require controlled manipulation of both acoustic and non-acoustic variables. We show that humanoid robots are well-suited as they enable parametric control of performance variables, reproducibility across trials, and the decoupling and recombination of auditory, visual, and interactive components. We illustrate the technical feasibility of this perspective through a case study of the WAseda Saxophonist Robot 5 (WAS-5), demonstrating reproducible control of acoustic and interaction variables that are prerequisites for future music-emotion experiments. Our work positions musical humanoid robots as a methodological platform that enables future controlled investigations of music-evoked emotions.
Read the full abstract

Multimedia · Cross-listed from cs.SD

Fourier is Frontier: Frequency-Aware Autoencoding for High-Fidelity Music Reconstruction

Kangdi Wang, Yusheng Dai and Jin Xu

Continuous-latent audio autoencoders form the backbone of latent music generators, yet decoders at high compression rates commonly exhibit three failure modes: high-frequency loss, phase incoherence, and stereo-image collapse. These share a structural root: waveform autoencoders lack an explicit frequency axis, leaving no handle for targeted per-band correction. Among five matched-budget representations, the complex STFT achieves the lowest full-band and high-frequency spectral distances, providing direct access to magnitude and phase at every bin. Building on this, we present ear-VAE2, a complex-spectral autoencoder with cross-channel interaction. Spec-SnakeBeta learns a periodic activation per frequency bin with frequency-dependent initialization, outperforming other activation variants while using fewer parameters than the fully independent variant. Duplex-Aware Refiner applies band-specific corrections to magnitude and phase following duplex theory of sound localization. On the 546-track Song Describer Dataset, ear-VAE2 achieves the best point estimates on five of seven reconstruction metrics. The Duplex-Aware Refiner reduces Mel Distance by 19.4% and uses ~45% fewer residual-output dimensions than the Unconstrained Refiner, while also lowering spectral distances, spatial-cue errors, and receiving higher ratings from professional engineers. The downstream generator using ear-VAE2 latents achieves better point estimates on all 12 automatic metrics.Demo page is available at https://eps-acoustic-revolution-lab.github.io/EAR_VAE2/.
Read the full abstract

Multimedia · Cross-listed from cs.CV

PEA-DPO: Perception-Enhanced Alignment Direct Preference Optimization for MLLMs Alignment

Jiawei Feng, Jiancan Wu, Xingyu Zhu et al. (+3 more)

Direct Preference Optimization (DPO) has emerged as an effective approach for aligning large language models (LLMs) with human preferences. However, its adaptation to multimodal settings remains unexplored. Through representational analysis, we identify a key limitation in multimodal preference optimization, which we term visual insensitivity: models often fail to distinguish between images and those with critical visual context removed. Our theoretical analysis further uncovers two manifestations of this problem, namely Across-Image Insensitivity and Within-Image Insensitivity. To address these challenges, we propose Perception-Enhanced Alignment DPO (PEA-DPO), a framework for multimodal LLMs alignment, which explicitly leverages visual preference signals to overcome visual insensitivity. We further provide a theoretical analysis demonstrating that PEA-DPO provably mitigates both failure modes. Empirical results demonstrate that PEA-DPO enhances sensitivity to visual context while preserving the language modeling capacity of the base model. Evaluations across three hallucination benchmarks using MLLMs of varying scales show that PEA-DPO effectively mitigates visual insensitivity, achieves stronger multimodal alignment, and substantially reduces hallucinations.
Read the full abstract

All authors Jiawei Feng, Jiancan Wu, Xingyu Zhu, Junkang Wu, Xiang Wang, Xiangnan He

Wednesday 19 August

Multimedia · Cross-listed from cs.GR

EMODY Flow: Emotion-Aware Audio-Driven Full-Body Motion Generation

Harsh Kumar Agarwal, Xavier Alameda-Pineda and Olivier Perrotin

Embodied conversational agents require synchronized full-body motion (body gestures and facial expressions) that aligns with speech and emotional state. Omni-modal large language models excel at multimodal understanding but produce only linguistic outputs, leaving a critical gap in embodied response generation. We identify and address a failure of emotion conditioning: like other conditional generators that under-use weak conditioning signals, a flow-matching model given both a rich audio embedding and a discrete emotion label suppresses the emotion, generating near-identical motion regardless of the specified emotion. We present EMODY Flow, a lightweight (around 35M parameters) flow-matching framework that attaches to a frozen Qwen-3 Omni model and reuses its internal Mimi audio-codecs to condition two parallel DiT generators - one for SMPL-X body pose, one for FLAME facial expressions. A training-time auxiliary emotion classifier restores emotion sensitivity by forcing generated motion to be emotion-identifiable. EMODY Flow sets a new state of the art on BEAT2 gesture quality, with FGD 0.302, Beat Correlation 0.853, and Diversity 24.62 - improving over the best prior results by 26%, 5%, and 62% respectively - and transfers to zero-shot facial animation on TFHP without domain-specific fine-tuning. Beyond these quantitative gains, the classifier yields clearly emotion-separated motion, which we demonstrate qualitatively through a multidimensional-scaling analysis of the generated gestures.
Read the full abstract