Search arXiv⌕ Search

SEARCH · Search arXiv

Results for “cs.MM”

Search indexed arXiv papers on artificial intelligence, large language models, computer vision and robotics. Read source abstracts and follow links to arXiv.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

7,800 records · Page 4Linked to original sources
arXiv2026

Agentic Artifact Creation: Systems, Evaluation, Principles, and Opportunities

Generative models can turn natural-language prompts into images, text, code, and other content, lowering the cost of producing drafts and components. Their practical impact increasingly depends on whether those pieces can become complete, dependable deliverables. This survey examines agentic artifact creation, which we define as stateful construction in which an AI system materially constructs or revises a deliverable and intermediate observations redirect later work. Functionally, the process links an operational representation of the artifact, a construction policy, and runtime verification whose feedback can redirect later actions. We reviewed 259 works available through August 20, 2026: 230 systems meeting this definition and 29 benchmarks of agentic artifact construction. We compare six artifact families, then analyze application settings and evaluation practice as separate dimensions. Across families, construction challenges reflect not only modality but also how tightly decisions are coupled and whether failures become visible while they remain repairable. Decomposition can reduce local complexity while increasing coordination and reassembly costs. Learned judges may add little independent evidence when they share the generator's preferences or blind spots. We formulate principles for keeping commitments and responsibility explicit, turning feedback into targeted repair, and revalidating affected state after change. We also identify opportunities for sustaining coherent, accountable control as artifacts, creator intent, and construction systems evolve. A curated paper list is available at https://github.com/GeminiLight/awesome-agentic-artifact-creation.

cs.MM↗
arXiv2026

Encore: Infinite Audio-Video Generation with Adaptive Signal Routing

Existing audio-video generation methods produce well-synchronized clips but are limited to short durations, while long-video generation methods extend duration through chunk-based iterative synthesis yet lack audio entirely. Generating long audio-video jointly is fundamentally harder than either task alone: each chunk must simultaneously maintain video temporal coherence, audio temporal coherence, and cross-modal synchronization, whose conditioning signals enter the model through different pathways. In this work, we present Encore for long-form synchronized audio-video generation. Our key insight is to factor this challenge into: (1) local continuity handled by iterative generation with explicit cross-chunk context propagation, and (2) global consistency enforced via reference audio-video signals with shifted position embedding. Building on this design, we propose Adaptive Signal Routing (ASR), which introduces learnable attention biases within self-attention and learnable residual scales on cross-attention outputs, enabling the model to adaptively modulate the influence of each conditioning signal. Trained end-to-end for joint audio-video generation, Encore also supports infinite-length audio-to-video and video-to-audio synthesis at inference by conditioning on the ground-truth modality throughout the denoising process. Experiments on our extended VerseBench for long audio-video evaluation demonstrate that Encore significantly outperforms existing methods in both generation quality and temporal coherence. Code and data for this paper are at https://github.com/shaohua-pan/Encore.

cs.MM↗
arXiv2026

Jailbreaking Text-to-Image Models Through Cracks: Navigating Heterogeneous Safety Filters via Multi-Agent Debate

Text-to-image (T2I) models remain vulnerable to jailbreak attacks that elicit Not-Safe-For-Work (NSFW) content, despite increasingly being guarded by heterogeneous, multi-layer safety stacks combining text filters, image classifiers, and cross-modal detectors. Existing jailbreak studies either optimize against individual filters or query the complete pipeline with aggregate feedback, making it difficult to identify the active constraint and adapt to conflicts across safety layers. In this paper, we introduce the Detection Surface, a unified geometric framework that characterizes the decision boundaries induced by heterogeneous T2I safety filters and their joint effect on the jailbreak search space. This formulation reveals that successful evasion is governed by a sparse and non-convex region shaped by cross-layer conflicts, where mutations that bypass one filter may increase exposure to another. Motivated by this analysis, we propose CRACK, a multi-agent debate framework for adaptive jailbreak search that decomposes jailbreak search into exploration, diagnosis, and arbitration. CRACK coordinates an Attack Agent, a Defense Agent, and a Judge Agent to iteratively generate prompt mutations, obtain layer-specific diagnostic feedback, and optimize mutation strategies through reward-guided refinement. Through repeated rounds of debate, CRACK adapts its search direction to the evolving cross-layer constraints while preserving the original harmful intent. Extensive experiments across multiple T2I models, datasets, and safety configurations show that CRACK achieves Attack Success Rates (ASR) of up to 99.63% under composite defenses, while requiring fewer queries than existing methods and maintaining semantic fidelity.

cs.AI↗
arXiv2026

The Missing Temporal Link: Temporal Context Routing for Script-Driven Audio-Video Generation

Joint audio-video generation models have made substantial progress in visual quality and audio-visual synchronization. However, they still provide limited control over when shot transitions occur and dialogue is spoken. This limitation constrains their application in script-driven content creation, where timing errors can undermine narrative coherence and the viewing experience. Current joint generators align video and audio representations on a shared temporal axis, yet the precise timing of shots and dialogue specified in a structured prompt is encoded only in the prompt's text representation and remains unaligned with the temporal coordinates of either modality. Consequently, video and audio may remain synchronized with each other while both fail to follow the script timeline. This mismatch motivates us to extend temporal alignment beyond video and audio to include the structured script. We therefore introduce Temporal Context Routing (TCR), which maps the script timing onto the shared temporal axis of video and audio generation and routes each prompt's guidance to the corresponding positions in both modalities. Compared with the baseline on 200 test scripts, TCR reduces Shot Boundary MAE by 96%, from 1.11 s to 0.042 s, and raises Dialogue Acc@0.5 s from 28.3% to 84.1%. TCR achieves these improvements while maintaining visual quality and audio-visual synchronization comparable to those of the baselines. A user study further shows that participants prefer TCR on all five evaluated dimensions.

cs.MM↗
arXiv2026

Scalable Neural Video Representation Compression

Scalable video coding (SVC) encodes a video into a layered bitstream consisting of a base layer and one or multiple enhancement layers, enabling decoding at different bitrate/quality/resolution operating points to accommodate diverse device capabilities and network conditions. Due to its practical flexibility, SVC has been incorporated into major video coding standards and has recently attracted growing interest for both scene-agnostic and scene-adaptive neural video codecs. Among the latter, Implicit neural representation (INR) based codecs achieve compression by overfitting a compact neural network to an individual video, offering fast decoding and competitive coding efficiency compared to scene-agnostic neural codecs. However, research on scalable INR-based compression remains in its infancy: these methods support scalable coding by introducing additional network layers, which couple the bitrate with the decoding complexity and also cannot achieve comparable performance with strong scalable/non-scalable codecs. In this context, this paper proposes S-NVRC, a scalable INR-based video codec that jointly supports fine-grained bitrate and decoding complexity scalability from a single embedded bitstream. It adopts a coarse-to-fine prefix for feature grids and a nested prefix for network layers, which scale bitrate and decoding complexity, respectively. The proposed S-NVRC spans a wide range of bitrate and decoding-complexity using a single encoding (training) and outperforms SHM 12.4 and the multi-layer VTM-20.0, by 43.7% and 5.6% in BD-rate on the UVG dataset, while also providing flexible complexity scalability. Implemented code will be provided.

eess.IV↗
arXiv2026

Identity-Driven Multimedia Forgery Detection via Reference Assistance

Recent advancements in "deepfake" techniques have paved the way for generating various media forgeries. In response to the potential hazards of these media forgeries, many researchers engage in exploring detection methods, increasing the demand for high-quality media forgery datasets. Despite this, existing datasets have certain limitations. Firstly, most datasets focus on manipulating visual modality and usually lack diversity, as only a few forgery approaches are considered. Secondly, the quality of media is often inadequate in clarity and naturalness. Meanwhile, the size of the dataset is also limited. Thirdly, it is commonly observed that real-world forgeries are motivated by identity, yet the identity information of the individuals portrayed in these forgeries within existing datasets remains under-explored. For detection, identity information could be an essential clue to boost performance. Moreover, official media concerning relevant identities on the Internet can serve as prior knowledge, aiding both the audience and forgery detectors in determining the true identity. Therefore, we propose an identity-driven multimedia forgery dataset, IDForge, which contains 249,138 video shots sourced from 324 wild videos of 54 celebrities collected from the Internet. The fake video shots involve 9 types of manipulation across visual, audio, and textual modalities. Additionally, IDForge provides extra 214,438 real video shots as a reference set for the 54 celebrities. Correspondingly, we propose the Reference-assisted Multimodal Forgery Detection Network (R-MFDN), aiming at the detection of deepfake videos. Through extensive experiments on the proposed dataset, we demonstrate the effectiveness of R-MFDN on the multimedia detection task.

cs.MM↗
arXiv2026

AVENUE: Audio-Video EditiNg Understanding and Evaluation

Audio-video (AV) editing aims to modify audio and video content according to a target prompt. Unlike single-modality editing, AV editing requires models to infer a modality-selective edit scope from the prompt alone: determining not only what should change, but also which modality should be preserved. Faithfully evaluating such models therefore requires both (i) benchmarks that span diverse edit types and modality categories, and (ii) evaluation that is itself modality-aware and sample-specific. However, existing AV editing benchmarks provide limited coverage of edit types and modality combinations, while current evaluation systems are often modality-blind and sample-agnostic, making it difficult to assess whether models faithfully preserve the unintended modality. To address these gaps, we introduce AVENUE, Audio-Video EditiNg Understanding and Evaluation, comprising two contributions: (1) a benchmark of 1,291 source clips and 7,957 editing instructions across audio-targeted, video-targeted, and AV-coupled edit types, curated and human-verified from VGGSound; and (2) a sample-specific, modality-aware evaluation framework that specifies, for each sample, both the intended change and the content that must remain intact. We evaluate representative AV editing models spanning three editing paradigms : joint, sequential, and separate, providing the first systematic analysis of modality-selectivity across paradigms. Our findings reveal a fundamental open challenge: when editing one modality, existing models frequently induce unintended changes in the other, regardless of paradigm. AVENUE provides a benchmark and modality-aware evaluation framework to drive progress toward more controllable AV editing models. Our dataset is publicly available on Hugging Face: https://huggingface.co/datasets/AVENUE-dataset/AVENUE.

cs.MM↗
arXiv2026

TempJail: Temporal Jailbreak Attacks against Image-to-Video Generation Models

In recent years, image-to-video (I2V) generation models have made remarkable progress in subject consistency and temporal coherence, enabling high quality video synthesis. However, these advances also introduce new safety risks. Existing studies mainly focus on jailbreak attacks involving single frame violations, while largely overlooking the temporal dimension unique to video generation models. In this paper, we investigate three attack scenarios and uncover a temporal vulnerability in I2V systems: unsafe semantics may emerge not from a single frame, but from semantic composition over time. We further identify two key challenges in such attacks: temporal abstraction and semantic camouflage. To address these issues, we propose TempJail, a novel temporal jailbreak framework for I2V systems. For temporal abstraction, we decompose a target malicious caption into an initial frame visual condition and a temporal text instruction. For semantic camouflage, on the image side we model semantic injection as controlled latent perturbation in diffusion sampling and introduce gradient guidance from pretrained encoders. On the text side, we rewrite the caption into an innocuous ``subject-action-scene'' template that bypasses safety filters while preserving temporal guidance. In the black-box inference phase, these two modalities jointly enable malicious semantics to be gradually triggered over time. Experiments on closed-source commercial models, including Kling, Seedance, Veo and PixVerse, show that TempJail improves attack success rate over prior state-of-the-art methods by 23.3\% under GPT-5.2 evaluation and 22.0\% under human evaluation. Our codes are available at \href{https://github.com/luqi-glory/TempJail}{GitHub}.

cs.CV↗
arXiv2026

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

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.

cs.MM↗
arXiv2026

Mind the Rift: Cross-Scale Coupling Mismatch for AI-Generated Video Detection

As AI video generators achieve cinematic realism, reliable detection becomes essential for safeguarding digital trust. We identify cross-scale coupling mismatch as a new forensic signal, where scale refers to the level of abstraction (semantic dynamics vs. pixel-level residuals): in natural videos, macro-level temporal dynamics and micro-level residual patterns are intrinsically coupled by the unified imaging physics pipeline, whereas AI generators, whose training objectives do not explicitly preserve this joint distribution, systematically violate this coupling. Detecting such mismatch is challenging because it requires independently extracting information at both scales while simultaneously quantifying their cross-scale relationship. We propose RIFT (Representation Inconsistency Forensics on Trajectories), an orthogonal forensic framework that addresses this through three interlocking components: a macro stream that builds a dynamic baseline of expected temporal evolution via differential geometry and persistent homology on learned manifold trajectories, a micro stream that acts as a sensitive forensic probe via steganalytic filtering and temporal modeling, and a coupling divergence module that measures the conditional dependency between the two streams. Gram-Schmidt orthogonality guarantees the information-theoretic validity of this measurement. Experiments on two benchmarks (VidProM, 120K videos, 7 generators; GenVidBench, 68K videos, 4 generators) demonstrate that RIFT achieves 99.33% and 99.72% F1-score respectively, with 97.87% unseen-generator detection rate in leave-one-out evaluation, while exhibiting encoder agnosticism: scaling from ViT-S/14 (22M) to ViT-L/14 (300M) changes F1 by less than 0.1%, and switching to a different encoder family (DINOv1) reduces F1 by only 0.73 pp. Code is available at https://github.com/Litsay/RIFT

cs.CV↗
arXiv2026

AOI-Net: Structural Face AOI-Guided Eye-Gaze Track Representation Learning for Autism Spectrum Disorder Detection

Eye-movement tracking has emerged as a promising non-invasive approach to Autism Spectrum Disorder (ASD) screening, with systematic differences in attentional allocation and revisit behaviors observed during socially interactive tasks. Existing computational methods typically characterize eye-movements using discrete gaze trajectories and fixation events, yielding representations dominated by short-range temporal dynamics and limiting models that primarily emphasize long-range dependencies. Meanwhile, gaze behavior is naturally organized across semantically meaningful Areas of Interest (AOIs), whose attention allocation and transitions provide important structural cues, yet their relationships are rarely modeled explicitly. To address these limitations, we propose a structural face AOI-guided Eye-Gaze Track Network (AOI-Net) that jointly models short-term temporal dynamics and AOI-level structural organization. A network gating mechanism adaptively integrates the complementary temporal and structural representations according to their contributions to gaze-behavior characterization. To mitigate the pronounced class imbalance commonly encountered between individuals with ASD and Typically Developing (TD) participants in clinical datasets, class-distribution-aware learning is further employed to facilitate discriminative embedding learning under skewed class distributions. Experiments on a unique and large-scale clinical eye-tracking database comprising eight stimulus subsets and more than 1,300 participants show that AOI-Net consistently outperforms state-of-the-art methods. The proposed framework also enables interpretable gaze-behavior modeling and provides a practical basis for scalable AI-driven ASD screening in real-world healthcare. The code is available at https://github.com/Zhanpei-ai/CIM-AOI-Net/tree/main/Code

cs.CV↗
arXiv2026

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

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.

cs.MM↗
arXiv2026

Data Diversity, Not Frequency Invariance: A Controlled and Self-Audited Study of Compression-Robust Deepfake Detection

Frequency features and compression-invariant representation learning are widely assumed to be key to deepfake detection that survives video compression. We test this with CAFRL - block-DCT and FFT-phase streams, compression-level-conditioned band attention, and adversarial (gradient-reversal) compression invariance - and report a controlled negative. Under a pre-registered protocol with capacity- and augmentation-matched controls, a plain EfficientNet-B0 on multi-quality data beat CAFRL as specified at every compression level on the FaceForensics++ test split, by 3.66 AUC points at CRF 40 (paired, single seed). A self-audit of our own negative found four defects biased against the frequency hypothesis, and pre-specified re-tests repairing all four showed the deficit to be a recipe artifact, not an architecture failure: the baseline recipe recovered 3.96 points over the matching shipped-recipe variant. The frequency path made no detectable difference: discriminative alone (standalone validation AUC 0.91-0.98 late in training) but of no marginal value under this fusion, at two feature widths of one 4.0 M trunk, every seed-pooled interval for the intra-dataset compression contrasts including zero; on the single held-out manipulation tested, the fair variants sat below the plain backbone. The adversarial branch, as specified, added nothing and degraded its own conditioning estimator; at the fair recipe it is untested. Robustness under single-pass H.264 re-encoding came instead from data diversity: real constant-rate-factor variants beat synthetic JPEG augmentation by 7.3 points (single runs, non-overlapping intervals). The evidence is FaceForensics++-family, GAN-era and single-codec. Match controls on training recipe as well as capacity, and buy compression robustness with codec diversity before architecture.

cs.CV↗
arXiv2026

Residual Maximin Share: Exact Finite-Agent Frontier, Sparse Extremizers, and Threshold Cuts

Residual maximin share (RMMS) is the largest share threshold that remains guaranteeable throughout dynamic allocation processes, even after previously allocated, lower-valued bundles are removed from the item pool. For additive valuations, recent density-balance analyses established finite-agent lower bounds comparing RMMS with the classical maximin share (MMS). In this paper, we prove that these finite-agent lower bounds are exact. Specifically, if $d_n$ denotes the largest odd integer at most $n$, the worst-case ratio satisfies $\inf_{M,v:\operatorname{MMS}>0}\frac{\operatorname{RMMS}(M,v,n)}{\operatorname{MMS}(M,v,n)}=\frac{2d_n}{3d_n-1}$. Consequently, the exact additive frontier forms consecutive odd-even plateaus and converges monotonically to $2/3$. We then investigate the combinatorial structure of extremal instances. While naive witnesses require $Θ(n^2)$ items, we construct an explicit three-valued family achieving the exact boundary with only linear support: $(5n-3)/2$ items for odd $n$ and $(5n-4)/2$ items for even $n$. Its low-valued block supports two exact partitions that simultaneously certify the MMS benchmark and the residual obstruction. By modeling these dual partitions as a bipartite transportation graph, we prove that this block attains the absolute minimum support $q+d-1=3q$. At minimum support, any two-valued filler is uniquely rigid up to relabeling. Finally, we establish structural characterizations of RMMS. A general min--max representation applies to all finite monotone valuations. For integer additive valuations, we prove that a threshold $T$ is residual self-feasible if and only if every subset cut satisfies a packing-covering condition. Because RMMS is pointwise maximal among residual self-feasible shares, these exact constants establish a tight limitation on the fairness guarantees achievable by share-based lone-divider algorithms.

cs.GT↗
arXiv2026

Discrete Gromov-Wasserstein Duality: Algorithms and Isomorphism Testing

The Gromov-Wasserstein (GW) distance provides a principled framework for aligning metric measure (mm) spaces based solely on their intrinsic structure. Its ability to identify isomorphic representations of distributions across spaces renders it valuable for comparing data where equality up to isomorphism occurs naturally such as in graphs or, more generally, distributions on graphs. Recently, a type of dual form for the GW distance between Euclidean distributions with the squared Euclidean or inner product costs was derived, spurring the development of new statistical and algorithmic results for this setting. This work furnishes a novel duality result for GW distances with and without entropic regularization that is applicable to all finitely supported mm spaces. Leveraging this result, we derive the sample complexity of empirical GW distances between finite mm spaces, as well as limit distributions under proper centering and scaling. Furthermore, we propose new algorithms for solving the regularized GW problem which are subject to formal convergence guarantees. These statistical and algorithmic advancements give rise to a principled and efficient framework for testing whether two distributions on the set of graphs with a fixed number of nodes are isomorphic based on samples.

math.ST↗
arXiv2026

A Data-Driven Multimodal Method for Early Detection of Coordinated Abnormal Behaviors in Live-Streaming Platforms

With the rapid growth of live-streaming e-commerce and digital marketing, abnormal marketing behaviors have become increasingly concealed and coordinated across heterogeneous modalities, challenging platform governance and early risk identification. We propose MM-FGDNet, a data-driven multimodal framework for detecting abnormal behavior in large-scale live-streaming environments from complementary temporal-evolution and group-structure perspectives. A cross-modal temporal alignment module maps video, text, audio, and user behavior into a unified temporal semantic space. A temporal fraud-pattern module captures the progression from weak early signals to abrupt outbreaks, while a cooperative manipulation module identifies coordinated interactions among organized user groups and automated accounts. Experiments on real-world multi-platform live-streaming e-commerce datasets show that MM-FGDNet outperforms representative baselines, achieving an AUC of 0.927, F1 of 0.847, precision of 0.861, recall of 0.834, and an Early Detection Score of 0.689, while reducing false alarms. Ablation studies validate the contribution of each module, and cross-domain experiments demonstrate stable generalization to new streamers, product categories, and platforms. These results indicate that MM-FGDNet provides an effective and scalable solution for proactive detection of coordinated abnormal behavior in live-streaming systems.

cs.SI↗
arXiv2026

Security and Privacy in the Musical Metaverse: Threat Analysis and Design Implications

The Musical Metaverse (MM) introduces immersive, real-time environments for collaborative musical interaction, characterized by ultra-low-latency constraints, continuous multimodal data streams, and heterogeneous devices. These properties create a distinctive security and privacy landscape that differs significantly from conventional XR or multimedia systems. This paper presents a multi-layer threat analysis of MM ecosystems, identifying key assets including live musical content, expressive interaction data, identity and session metadata, and intellectual property. Threats are analyzed across network, application, data/AI, device, intellectual property rights, and social layers, with particular attention to risks arising from expressive and neurophysiological data, which enable inference, re-identification, and potential privacy violations. We describe a stakeholder-driven survey involving 14 participants from 13 organizations, revealing that neurophysiological data leakage and real-time stream disruption are perceived as the most critical risks, followed by intellectual property infringement and avatar impersonation. We further evaluate the suitability of existing security protocols under strict latency constraints, showing that conventional approaches such as TLS over TCP are often incompatible with real-time musical interaction, while lightweight, stream-oriented mechanisms (e.g., SRTP, DTLS) provide a more suitable balance between security and performance. Based on these findings, we derive a set of design guidelines for MM systems, emphasizing latency-aware security, differentiation of interaction paths, data minimization, and edge-centric processing. The results support a security-by-design approach that enables trust and compliance without compromising real-time performance.

cs.CR↗
arXiv2026

Two-Stage Multi-Modal Fusion with Adaptive Alignment for Action Quality Assessment

Action Quality Assessment (AQA) aims to evaluate how well a person performs a movement, which is essential in applications such as sports scoring, skill assessment, and healthcare. However, unimodal approaches often struggle to capture subtle cues of movement quality in real-world settings. Although multi-modal inputs provide complementary information, existing methods still face two major challenges: heterogeneous modalities often lead to cross-modal misalignment and unstable fusion, and reliable multi-modal annotation is costly, resulting in limited dataset diversity. To address these challenges, we propose DualAlign, a two-stage multi-modal fusion framework with adaptive alignment. The framework first constructs a coherent visual representation by maximizing shared structural information across RGB video, optical flow, and skeleton modalities. Textual semantics are then incorporated after visual stabilization, allowing high-level descriptions to complement rather than distort the underlying visual manifold. To evaluate the framework under realistic multi-modal conditions, we introduce MM--JDM, a movement-quality assessment dataset integrating RGB videos, optical flow, skeleton sequences, and structured text. MM--JDM naturally exhibits modality noise, class imbalance, and label scarcity, making it a challenging benchmark for studying multi-modal fusion and alignment. Extensive experiments show that DualAlign improves average correlation on MM--JDM by 21.16% over the state-of-the-art methods and achieves gains of 3.53% and 5.95% on the RG and Fis-V benchmarks, respectively. DualAlign also remains robust under missing-modality and label-scarce conditions.

cs.CV↗