Multimodal

Lecture videos distribute knowledge across speech, slide text, diagrams, equations, and presentation order, which transcript-only retrieval does not fully preserve. This paper presents an evidence-grounded multimodal pipeline that transcribes lectures, selects semantic anchors, applies optical character recognition (OCR), and uses a vision-language model to extract only concepts and typed relationships supported by transcript, OCR, or visual evidence. Mentions are validated and canonicalized into a provenance-rich knowledge graph. On three neural-network lectures, the pipeline processed 3,118 frames, 756 transcript segments, and 559 anchors. It retained 1,022 concept and 312 relationship mentions, yielding 172 canonical concepts and 282 relationships with 90.38% endpoint coverage. A preliminary three question retrieval test achieved 100% top-1 and top-3 accuracy and 100% mean top-5 recall. The contribution is an auditable construction method rather than a state-of-the-art performance claim.

Multimodal large language models (MLLMs) achieve strong performance across diverse vision-language tasks, but their efficiency is limited by the cost of processing numerous visual tokens. Visual token pruning can reduce this cost, but requires accurate token importance estimates. Recent studies have demonstrated that text-to-vision attention from middle language model layers can effectively guide visual token pruning, typically using attention from a predefined middle layer to select the visual tokens to retain. Two problems therefore remain. First, our analysis shows that the layer whose attention is most responsive to the question varies substantially across samples, making a fixed layer suboptimal. Second, obtaining attention from the appropriate middle layer requires processing numerous visual tokens through several language model layers, by which point considerable computation has already been spent. To address both problems, we propose Middle-layer Attention Prediction (MAP), which uses Question Contrastive Teacher Selection to identify a sample-specific teacher layer by contrasting attention under the original and reference questions, and distills attention from the selected layer into a lightweight predictor that estimates visual token importance from multi-modal input features. During inference, MAP combines the predicted importance scores with a diversity criterion to prune visual tokens before the first language model layer. Thus, MAP requires no attention maps for pruning and remains compatible with existing inference acceleration techniques. Across ten benchmarks on LLaVA-NeXT-7B, MAP retains 97.5% of the unpruned model performance with only 5.56% of the visual tokens, yielding a 3.09x end-to-end speedup.

We study contextual dynamic pricing with arbitrary covariate sequences and bounded, possibly nonbinary purchase quantities. Demand follows a semiparametric surplus-index model with an unknown linear valuation parameter and an unknown Hölder-smooth response. We impose neither concavity nor strong unimodality on revenue and allow nonunique optimal prices. We develop a pilot-corrected layered decision-partitioning policy that combines directional pilot estimation, local polynomial learning, predictable data assignment, and global action elimination. Pilot correction removes the first-order effect of valuation-parameter error, while permanent labels enable concentration under adaptive sampling. The policy attains the minimax smoothness-dependent horizon rate up to logarithmic factors; a matching lower bound already holds for a constant-context binary-demand subclass.

Multimodal large language models have demonstrated strong defect recognition capability in industrial anomaly detection. However, in lithography review, merely determining whether an image contains a defect is insufficient for engineering inspection; models must also understand defect morphology, spatial location, and the potential causes supported by visible evidence. To this end, this paper proposes LDU-Bench, a multi-task multimodal benchmark for lithography defect understanding. Constructed from real lithography and integrated-circuit review images, LDU-Bench decomposes the review workflow into four independent tasks: defect triage, morphology recognition, coarse localization, and image-conditioned cause analysis. It systematically evaluates models using task-level metrics, diagnostic readouts, and the Lithography Closure Score (LCS). Experimental results show that although existing MLLMs can perform defect triage relatively reliably, this ability does not stably transfer to downstream review stages. Morphology alignment, effective localization, and evidence-to-cause mapping remain the major bottlenecks. Further diagnostics indicate that this capability break is not a fluctuation of a single metric, but reflects insufficient structured understanding across semantic levels. Overall, LDU-Bench provides a quantifiable and diagnostic unified platform for evaluating the usability, failure points, and capability boundaries of industrial MLLMs in lithography review chains.

What is music style? Though often described using text labels such as "swing," "classical," or "emotional," the real style remains implicit and hidden in concrete music examples. In this paper, we introduce a cross-modal framework that learns implicit music styles from raw audio and applies them to symbolic music generation. Inspired by BLIP-2, our model leverages a Querying Transformer (Q-Former) to extract style representations from a large, pre-trained audio language model (LM), and further applies them to condition a symbolic LM for generating piano arrangements. We adopt a two-stage training strategy: contrastive learning to align auditory style with symbolic expression, followed by generative modeling for music arrangement. Our model generates piano performances jointly conditioned on a lead sheet (content) and a reference audio example (style), enabling controllable and stylistically faithful arrangement. Experiments demonstrate the effectiveness of our approach in piano cover generation, style transfer, and audio-to-MIDI retrieval, achieving substantial improvements in style-aware alignment and music quality.

Large Multimodal Models (LMMs) have achieved remarkable success on images and short videos, yet scaling them to long videos remains challenging due to frame-centric tokenization and limited context windows. 3D geometry provides a natural compression mechanism for visual streams: depth and camera pose enable observations from multiple views and time steps to be fused into a persistent, world-aligned representation. While recent 3D LMMs leverage geometry-aware representations to improve spatial reasoning, they continue to lag behind specialist 3D perception systems on grounding and segmentation tasks. We argue that a key limitation is geometry-aware decoding: existing methods communicate 3D predictions through language tokens, proposal selection, or lightweight grounding queries, creating a bottleneck between language reasoning and dense geometric prediction. Building on these insights, we introduce Qwen-3D, a geometry-aware LMM that compresses visual information within the Qwen backbone using multi-view geometric cues, enabling efficient long-horizon visual reasoning over static scenes. Qwen-3D augments visual tokens with 3D Rotary Positional Embeddings, allowing attention to operate directly in 3D scene space rather than across independent image frames and thereby facilitating scalable cross-view and temporal reasoning. To bridge language and geometry, Qwen-3D incorporates a query-based segmentation decoder that grounds language directly in the underlying 3D scene representation, unifying referential grounding, instance segmentation, and visual question answering across both images and videos. Across a diverse set of benchmarks, Qwen-3D surpasses existing 3D LMMs and outperforms several large proprietary 2D models. Notably, Qwen-3D achieves these improvements while maintaining strong performance on standard 2D vision-language benchmarks by jointly training on 2D and 3D data.

Multimodal supervised learning seeks to leverage multiple heterogeneous data sources to improve predictive performance. A central challenge is determining the fusion granularity across modalities: over-integration may amplify noise while under-integration fails to exploit cross-modal dependence. Existing approaches rely on pre-specified fusion architectures, from early to late fusion, that may not adapt to the underlying dependence structure among modalities. We propose DAIF, a data adaptive intermediate fusion framework that combines random matrix theory and non-parametric dependence measures to learn fusion structure directly from data. We operate under a Bayesian multimodal factor model where the prior on the latent factors determines the cross-modal dependence. Our method clusters modalities based on estimated intermodal dependence, then performs clusterwise empirical Bayes estimation of the priors. These estimated priors are used to construct denoisers within an approximate message passing (AMP) framework, yielding denoised low-dimensional features that borrow strength across related modalities while preserving modality-specific signal. The resulting embeddings are used for downstream supervised prediction. We evaluate the framework through simulations under varying dependence structures and signal regimes, comparing against several benchmark methods, and demonstrate its practical utility on two multimodal datasets, namely a trimodal TEA-seq dataset (Swanson et al., 2021) and TCGA-BRCA dataset (Goldman et al., 2020). In the first example, we predict the expression level of a T-cell differentiation marker protein and in the second case we analyze patient survival prediction based on multimodal information. Our method competes with or outperforms the state-of-the-art techniques in both prediction problems, demonstrating its versatility across diverse supervised learning tasks.

Automated pain assessment in real clinics is limited by scarce clinically grounded facial video data with weak labels (often sequence-level self-report) and by the fact that pain cues can be subtle or near-neutral in RGB, while thermal and depth signals are informative yet impractical to deploy routinely. To address these challenges, we propose ReMiX-MAE (Reconstructing Missing Channel Cross-Modal Masked Autoencoder), a self-supervised multimodal masked pretraining framework that learns transferable facial representations from synchronized RGB, thermal, and depth videos and explicitly trains robustness to missing modalities, enabling RGB-only deployment. To fill the gap of clinically grounded facial pain data with video-level self-report and longitudinal treatment trajectories, we collect the Sympathetic Mediated Pain (SMP) dataset with paired pre- and post-recordings across multiple visits. Under RGB-only deployment, we evaluate ReMiX-MAE using both direct feature extraction and pseudo-multimodal features decoded from RGB. ReMiX-MAE consistently outperforms an RGB-only masked autoencoder baseline on SMP, with pseudo-multimodal features providing additional gains in the challenging five-class setting. Across external datasets, ReMiX-MAE further shows more robust and label-efficient transfer than RGB-only baselines, highlighting its advantage in data-limited clinical settings.

Audio-driven video generation (A2V) has achieved promising progress in synthesizing temporally coherent and audio-visually aligned videos, yet its inference remains expensive due to the iterative denoising process of diffusion models. Existing caching methods mainly exploit temporal redundancy in visual features while overlooking the cross-modal alignment of A2V, where audio drives visual generation with highly non-uniform temporal importance. In this paper, we identify two levels of misalignment in existing A2V caching methods: temporal-semantic and computation-storage misalignment. To address them, we propose EchoCache, an energy-guided cross-modal caching framework for efficient A2V generation. EchoCache leverages audio time-frequency energy as a saliency anchor to guide latent-level cache updates and further introduces a dynamic timestep-latent caching mechanism with quantized cache management for joint efficiency and memory optimization. Extensive experiments on mainstream A2V models show that EchoCache consistently improves the latency-quality trade-off while preserving generation quality and audio-visual consistency. In particular, on Wan2.2-S2V over the EMTD benchmark, EchoCache achieves a 2.46x speedup with the best overall performance. Code is available at https://github.com/IF-LAB-PKU/EchoCache.

Scientific formulae are a fundamental component of scholarly communication, yet their dual nature -- as structured syntax and carriers of semantics -- remains underexplored in scholarly information retrieval. Although prior studies show that jointly modeling syntactic and semantic modalities improves retrieval performance, the relationship between their underlying representations has not been systematically investigated. In this work, we empirically study cross-modal correspondence between formula syntax and semantics. We find that their native representation spaces exhibit extremely weak observable correspondence despite strong latent correlation, indicating a substantial representation mismatch between the two modalities. We further evaluate whether this mismatch can be reduced using standard representation learning and alignment techniques. We represent syntactic structure using graph-based encoders and semantic information using text-based encoders, then apply contrastive learning to induce a shared representation space. Results show that the learned alignment substantially improves cross-modal retrieval, suggesting that explicit representation learning can recover correspondence absent from the original representation spaces.

Deploying vision-language models (VLMs) for safety-critical spatial reasoning on resource-constrained autonomous driving platforms requires both compact model size and reliable metric grounding. We present MoRAL (Multimodal Reasoning for Autonomous Language Models), a two-stage fine-tuning pipeline that teaches Cosmos-Reason2-2B to first read a physics-encoded Bird's Eye View (BEV) representation and then reason over it for driving decisions. The BEV image encodes LiDAR metric distance as color bands, object class as cluster morphology, and radar Doppler velocity as directional wedge overlays, externalizing spatial perception into the input image so that no learned 3D backbone is required at inference. Stage 1 fine-tunes the vision encoder on 60,000 grounding records; zero-shot baselines produce no parseable BEV outputs, confirming the vocabulary requires explicit training. Stage 2 fine-tunes the full model (52M parameters, 2.4% of total) on 57,696 chain-of-thought records generated by Cosmos-Reason2-8B as teacher, spanning eight driving question types. On 2,304 held-out nuScenes frames evaluated by Gemma 4 (31B) calibrated against human review, MoRAL wins seven of eight question types over a zero-shot 8B baseline despite using four times fewer parameters, with the largest margins on question types requiring structured multi-step physics reasoning. Emergency braking recall improves from 10.8% to 47.8%, output degeneration falls from 94.1% to 20.8%, and the full pipeline fits a consumer 8 GB GPU at 42 tok/s without quantization. These results establish a reproducible foundation for compact, physics-grounded VLM reasoning on mobile edge platforms.

Longitudinal multiparametric MRI is central to follow-up imaging in oncology, yet real-world clinical data are characterised by missing sequences, heterogeneous acquisition protocols, and varying spatial resolutions across time points. We propose a patient-specific conditional implicit neural representation (INR) that models multimodal longitudinal MRI as a continuous function of world coordinates, time, and modality conditioning. The model is trained with stochastic modality dropout to handle incomplete data, and its continuous coordinate-space formulation enables both spatial and temporal interpolation without resampling to a fixed voxel grid. A self-consistency-based confidence estimator is derived from cross-modal reconstruction performance at inference time. We evaluate the framework on longitudinal MRI from paediatric brain tumour patients, demonstrating statistically significant improvements over linear interpolation for T1CE and FLAIR (p < 0.05), with mean MS-SSIM of 0.95 $\pm$ 0.02 for T1CE. Predicted confidence correlates strongly with true reconstruction quality (Pearson r up to 0.996), suggesting reliable deployment potential in heterogeneous clinical settings.

We propose Sen-Cap, a Sensor-Flexible and Noise-Resilient 3D human motion Capture framework that integrates multi-modal data from LiDAR and camera. While multi-modal sensors provide richer information than single-modal sensors, existing approaches still suffer from two core challenges. First, multi-modal alignment/matching across arbitrarily deployed sensors is typically handled by explicit calibration, which propagates errors under changing viewpoints and in turn constrains deployment to fixed, highly overlapped layouts. Second, prior methods degrade under severe noise or partial sensor failures, which are common in real-world environments. To address these challenges, Sen-Cap introduces a Unified Across-Sensor Motion Estimator that reconstructs local pose and shape in a human-centric space without calibrations between sensors, supporting a flexible number of sensors, as well as a Noise-Resistant Trajectory Tracker that maintains robustness under severe point cloud noise through iterative refinement. These sensor-flexible and noise-resilient features make Sen-Cap more practical in real-world deployment. Notably, operating in real time, Sen-Cap achieves state-of-the-art performance on major metrics on Human-M3 and FreeMotion, as well as strong cross-domain performance on LiDARHuman26M and RELI11D. This combination of flexibility and robustness opens new opportunities for motion capture in real-world scenarios, e.g. sports analytics, field robotics, and large-scale immersive environments.

Scientific poster construction compresses a long multimodal paper into a readable, editable canvas. Existing systems hide request-level failures by scoring only completed outputs; direct image generation is not element-editable, while coding-agent workflows are costly. PosterMELD is a template-conditioned multi-agent pipeline: capacity-aware slots guide writing before rendering, and deterministic gates plus vision-language model (VLM) review route failures to bounded repair. Each accepted request exports editable PowerPoint (PPTX) and Portable Network Graphics (PNG) artifacts; explicit design controls yield same-paper variants. Across 621 papers, Print-Ready Rate (PRR) counts requests passing geometric, readability, asset-integrity, and obvious-factual-error checks, with native editability reported separately. A frozen VLM assigns conditional Craftsmanship-Harmony-Expressiveness (CHE) scores to print-ready outputs. PosterMELD attains 81.3% PRR, 3.4 times P2P's rate and 5.2 times PosterGen's, and the highest conditional CHE among generated methods with multiple print-ready outputs. Native editability and explicit design controls are retained at a mean cost of USD 0.38 per request, 3.5% of Codex+Skill's. Code and resources are available at https://github.com/Shannon4Science/PosterMELD.

Multimodal representation learning is a cornerstone of modern AI. By encoding multimodal queries and targets into vectors, it powers industrial search and recommendation and underpins modern agents. Real-world platforms with complex modalities and massive-scale content, such as Douyin, Xiaohongshu, and YouTube, demand both efficiency under billion-scale indexing and fine-grained discrimination for hard matching. Existing MLLM embedding models rarely satisfy both. Contrastive models are efficient but rely on pair-level supervision too coarse for fine-grained distinctions, while CoT-based models improve discrimination through explicit generation impractical to serve online. We present Douyin Multimodal Embedding (DME), a model trained in two stages to combine both strengths. Stage 1 performs large-scale contrastive pre-training that establishes a unified multimodal embedding space with broad modality and task coverage. Stage 2 supplements semantic sufficiency, the property that an embedding is grounded in retrieval-relevant evidence and preserves fine-grained counterpart-side semantics, via two mechanisms. Evidence-Grounded Typed Latent Reasoning organizes retrieval evidence through hidden-space latent reasoning, and Cross-Conditional Reconstruction enforces counterpart-side semantics through cross-directional autoregressive reconstruction. Both act only during training and add only marginal query-side overhead, so DME serves as efficiently as a standard contrastive encoder. On MMEB-v2, DME reaches state-of-the-art results at comparable scales for its 2B and 9B variants (74.8 and 78.4), with especially strong video and visual-document tasks. In production, DME delivers a 2.92% relative gain on Douyin's in-house offline evaluation set, is deployed across Douyin scenarios such as generative, image, and AI search, and yields a 0.1% Lifetime (LT) gain in online A/B testing on Douyin search.

Vision-language models (VLMs) exhibit strong generalization across multimodal tasks but remain vulnerable to adversarial perturbations. Existing attacks typically follow single-trajectory gradient optimization or task-specific objectives, limiting search-space exploration and cross-task transferability. We propose an evolutionary-computation-guided cross-modal attack framework for unified VLMs. The framework adaptively searches both textual and visual spaces. On the textual side, it evolves hard negative semantic embeddings around the source-category representation to provide diverse cross-modal repulsion. On the visual side, it maintains a population of object-region perturbations and combines momentum-based gradient updates with evolutionary selection, mutation, and crossover to more reliably explore multiple feasible trajectories. Jointly optimizing semantic negative guidance and localized perturbations generates adversarial examples that consistently shift source-object semantics toward target categories across vision-language tasks. Theoretical analyses show that the co-evolutionary search preserves perturbation feasibility, prevents degradation of the best observed fitness, and increases the probability of reaching high-margin adversarial regions compared with single-trajectory optimization. Experiments on Florence-2, OFA, and UnifiedIO-2 demonstrate strong overall attack performance across image captioning, object detection, region categorization, and object localization. Ablation studies further verify the complementary effectiveness of text-side semantic evolution and image-side perturbation evolution, as well as the framework's efficiency and cross-task transferability.

Deep multimodal fusion for object detection has demonstrated good performance through mining modal characteristics. However, existing feature-level fusion methods mainly weigh between two modalities and unify them in a unified representation space. This can lead to overfitting or over-specialization of the statistical properties of a single modality within a dual-backbone architecture. This paper proposes an Attention-Driven Complementarity Resampling framework for robust improvement of cross-modality object detection. Based on a shared channel spatial attention mechanism, we first introduce the semantic mask exchange to actively mix the boundaries of the modalities during the training phase, forcing the backbone network to learn generalized features without relying on fixed modal labels. Then we propose a learnable channel competition to sample and aggregate features in a channel-wise and learnable way. Our experiments on multiple datasets demonstrate that the proposed method is effective and yields competitive results among existing state-of-the-art approaches. The source code is provided in the supplementary material.

Recent advances in unified multimodal models have significantly improved text-guided image editing abilities. In particular, models such as Nano-Banana-Pro and GPT-Image-2 demonstrate emerging capabilities in multi-source image editing (MIE), including tasks such as object synthesis, person-background composition, and cross-image style fusion. However, existing benchmarks and image editing assessment (IEQA) methods remain primarily focused on single-image editing tasks and largely overlook the more challenging setting of MIE. This highlights the urgent need for a comprehensive and human-aligned benchmark for MIE. To this end, we introduce MIE-Bench, the first large-scale multiple image editing benchmark with fine-grained human preference annotations. Specifically, MIE-Bench includes 3,000 editing instances across 16 tasks, each involving more than two source images and an editing prompt, together with 36K edited images produced by 12 state-of-the-art editing models and over 108K mean opinion scores (MOSs) covering visual quality, instruction following, and attribute preservation. Based on MIE-Bench, we propose MIEScore, a multimodal large language model (MLLM)-based evaluation model enhanced with skill optimization and multi-dimensional supervised fine-tuning, to provide human-aligned feedback for MIE. Extensive experiments show that MIEScore achieves state-of-the-art performance in aligning with human preferences and generalizes well across other IEQA datasets. Both the dataset and the model are available at https://github.com/IntMeGroup/MIEScore.

Access to holistic, multimodal data improves the performance of Artificial Intelligence (AI) in medical classification tasks compared to utilizing single modalities or data sources. However, the inherent heterogeneity and complexity of clinical real-world data pose significant challenges to structured data analysis and AI application. This heterogeneity includes missing values, multiple time points, diverse modalities, and inconsistent formats and semantics. Data harmonization prior to data integration tackles this challenge but remains resource-intensive and error-prone, limiting the scalability and reproducibility of holistic, AI-driven decision support on clinical real-world data. We therefore propose PatTree, a graph-based, holistic representation of patients that can be derived from real-world clinical data through the automated structuring of multimodal clinical data. PatTree enables early-stage data integration without relying on pre-standardized inputs. While representing heterogeneous clinical data within a unified knowledge graph, PatTree preserves the semantic relationships between data elements across modalities and data sources, facilitating interoperability and machine-interpretable data access. Using a subset of the ADNI-1 cohort (n = 763), we demonstrate that classification of patients is directly feasible on PatTree reaching state-of-the-art classification performance. In the three-class classification task distinguishing Alzheimer's disease, mild cognitive impairment, and cognitively normal individuals, we achieve a balanced accuracy of 98.5% and an F$_1$ score of 0.987 on the held-out test set. Our results show that assumption-free, automated structuring of multimodal medical data can serve as a scalable foundation for clinical AI pipelines bypassing tedious data preparation and standardization.

Visual token pruning reduces the inference cost of multimodal large language models, but a fixed token ratio is poorly matched to text-rich inputs. In OCR-centric tasks, decisive evidence can be a small number, label, or field whose relevance is specified by the question; indiscriminate pruning can erase that evidence while retaining visually salient but irrelevant regions. We present ET-Prune, a training-free framework that casts pruning as evidence allocation. It derives question-conditioned evidence from a decoder-side partial query-key block, safeguards text-like spatial regions, and converts evidence uncertainty and density into a sample-specific token floor. Three progressive middle-layer events then move the sequence toward this budget, retaining more tokens for diffuse or text-dense evidence and pruning concentrated evidence more aggressively. At the observed point estimates from one deterministic pass per configuration, ET-Prune leads or ties among pruned methods in all six backbone-benchmark comparisons at roughly half tokens. On OCRBench-v2, it leads the strongest pruned baselines by 1.80 and 0.68 percentage points on Qwen3-VL-8B and InternVL3.5-8B, respectively, while retaining about half of the visual tokens; on MMBench v1.1, it reaches 0.8467 circular exact-matching accuracy versus 0.8437 for Vanilla at 54.45% average visual-token retention. These results show a favorable observed quality-cost trade-off for evidence-aware dynamic budgeting in text-rich multimodal inference.

Multimodal large models are increasingly used to generate scalable vector graphics (SVG), but reliable evaluation remains underexplored. Existing protocols are often code-centric or borrow raster-image metrics after rendering SVGs, which fail to reflect human perception and overlook SVG-specific qualities such as geometry and spatial composition. We introduce SVGEval, a vision-grounded multimodal benchmark for human-aligned SVG quality assessment. SVGEval explicitly incorporates visual renderings to evaluate whether models can judge the rendered outcome rather than only inspect SVG code, and provides high-quality annotations obtained via multi-round human labeling with expert refinement. Systematic evaluations across representative multimodal models reveal a clear gap: models perform relatively well on semantic alignment and aesthetics, yet struggle on geometry- and layout-related judgments. Building on SVGEval, we train an explainable SVG quality scorer that outputs multi-aspect scores with textual rationales. Ablations show that explicit visual grounding and reasoning supervision are crucial, especially for spatial and geometric assessment. SVGEval offers a reliable testbed and practical scorer for evaluating and improving SVG generation in the era of multimodal models.

Multimodal Large Reasoning Models (MLRMs) have achieved strong performance on tasks requiring visual understanding and multi-step inference. However, as reasoning trajectories grow, models may become less effective at using information established earlier in the context, increasing the risk of reasoning errors. Existing approaches primarily address this problem by sustaining visual grounding throughout reasoning. However, reasoning also transforms visual observations into task-specific relations, constraints, and intermediate conclusions whose influence may weaken over long trajectories. Our attribution analysis suggests that correctness is not consistently separated by image attribution alone, but is more closely associated with whether trajectories retain and integrate such reasoning-derived information across stages. Motivated by this, we introduce TRAM (TRajectory-derived Auxiliary Memory), a training-free method that augments standard decoding with an auxiliary memory pathway derived from the model's own reasoning trajectory. TRAM consolidates completed reasoning into a compact latent memory, updates it online through fast and slow recurrent streams, and feeds it back into selected decoder layers through a lightweight residual pathway. Experiments across four MLRM variants on eight benchmarks show that TRAM improves performance over vanilla decoding on mathematical, scientific, and general visual reasoning tasks without additional training.

Phenotypic drug discovery enables the discovery of functional relationships between molecular structures and cellular responses. However, existing multimodal representation learning methods often optimize cross-modal alignment without considering the intrinsic organization of chemical space, resulting in distorted molecular representations and loss of structural information. We propose \textbf{PhenMol}, a structure-preserving framework for phenotype-aware molecular representation learning. PhenMol disentangles molecular and cellular representations into shared and private components, enabling phenotype-guided alignment while preserving chemical structures through a dedicated molecular branch. This design integrates cellular phenotype information without disrupting molecular neighborhood organization. Experiments on approximately $3.04 \times 10^{4}$ molecule--cell morphology pairs demonstrate that PhenMol improves molecular property prediction across 270 bioactivity tasks, molecule--phenotype retrieval, and clinical trial outcome prediction. Moreover, ECFP4-based structural analysis shows that PhenMol better preserves molecular neighborhoods and reduces embedding distortion compared with existing multimodal alignment methods. These results highlight the importance of structure-aware constraints in multimodal molecular representation learning and provide an effective approach for integrating cellular phenotypes with chemical knowledge for drug discovery.

Machine unlearning offers a promising approach to remove unsafe content from Multimodal Large Language Models (MLLMs), yet ensuring the precision of unlearning remains a persistent challenge. One reason is that current MLLM unlearning evaluation paradigms suffer from a critical blind spot: they assess model utility through benchmarks whose representations are distant from the forget set, failing to capture knowledge holes---severe degradation on benign adjacent inputs. To probe knowledge holes in unlearned MLLMs, we construct a benchmark that captures unintended degradation on benign inputs sharing generic patterns with the forget set, and confirm through controlled experiments that they are a systematic consequence of commonly used approaches. Furthermore, to bridge this gap, we propose Selective Protection with Anchored Regularization, which protects generic patterns via anchored activation filtering while reinforcing them through entity-abstracted enhancement. Our experiments on SafeEraser demonstrate that SPAR recovers over 98% of vanilla response quality compared to below 50% for standard baselines---while achieving 0.00% attack success rate and competitive model utility. These results underscore the necessity of more fine-grained evaluation for trustworthy MLLM unlearning.

Universal Multimodal Embeddings (UMEs) aim to unify various modalities and tasks into a shared representation space. In recent years, this field has witnessed substantial progress driven by the development of Multimodal Large Language Models (MLLMs). However, a crucial capability, visual identity discrimination, remains underexplored in existing UME methods, despite its critical role in a wide range of tasks, including instance retrieval, re-identification, and identity preservation in AI-generated content. To bridge this gap, we propose a unified formulation for visual identity discrimination~(VisID) and introduce $\textbf{MVEB}$ ($\textbf{M}$ultimodal $\textbf{V}$isual Identity $\textbf{E}$mbedding $\textbf{B}$enchmark), a large-scale benchmark curated from both real-world and synthetic datasets to support evaluation and training. Furthermore, we present a simple yet effective learning framework that jointly optimizes general multimodal and visual identity representations through a carefully designed identity-aware sampling mechanism. Extensive experiments demonstrate that our approach successfully endows UMEs with strong identity discrimination capability and maintains competitive general multimodal performance. We believe this work not only illuminates a critical yet neglected capability, but also takes a step toward more holistic universal multimodal embeddings. Code and data are available at \href{https://chrisclear3.github.io/MVEB}{MVEB}.

Token compression in OmniLLMs is typically posed as a single saliency-ranking problem: score each multimodal token, keep the top-K. We argue this abstraction is mis-specified. The same attention score simultaneously decides two things: how much retained capacity each modality receives, and which tokens within a modality are kept. A shared top-K rule therefore inherits this audio-favoring allocation prior, spending retained capacity on audio before video tokens have a chance to compete. We propose Macer, a training-free compressor that first assigns explicit audio and video budgets, then performs allocation-normalized ranking within each modality at modality-specific shallow layers. Macer significantly reduces token cost while preserving accuracy across audio-grounded, audio--video joint, visual-dominant, and video-centric benchmarks. At 25 % retention, Macer preserves 98.7 % of full-token performance on Qwen2.5-Omni-7B and 97.3 % on Qwen2.5-Omni-3B. On Qwen2.5-Omni-7B, this 25 % setting reaches OmniZip-level performance at 45 % retention while using lower FLOPs. On OmniVinci-9B, the same allocation-before-ranking principle improves over shared top-K ranking by up to 12.9 points.

We present our ImageCLEF 2026 Multimodal Reasoning system for the Visual Multiple Choice Question Answering (Visual MCQ) and Visual Open Question Answering (Visual OpenQA) subtasks. The challenge requires reliable reasoning over multilingual educational and scientific images with dense text, diagrams, charts, tables, formulas, and units, while enforcing strict answer formats. Our central finding is that robust output control is as important as model choice. For Visual MCQ, we replace fragile free-form generation with direct candidate label scoring from vision-language model logits, then combine complementary runs through score fusion and voting. For Visual OpenQA, we use image enhancement, concise final answer prompting, deterministic decoding, and targeted post-processing to remove reasoning traces and formatting artifacts. Without task-specific model training, our official submissions achieved third place in Visual MCQ with 0.7108 accuracy and first place in Visual OpenQA with 0.6488 COMET, 0.1391 BLEU, 0.2762 ROUGE L, and 0.2383 METEOR. The results highlight the practical value of inference engineering: careful scoring, ensembling, prompting, and cleanup can turn strong VLMs into reliable competition systems.

Despite the progress of multimodal large language models (MLLMs), they continue to exhibit deficiencies in visual perception. Following visual instruction tuning, internal MLLM representations rapidly deviate from their original semantic states during inference, causing severe information degradation. While existing methods attempt to leverage external vision foundation models (VFMs) to align internal representations, we find that direct alignment with VFMs enhances visual semantics but fails to mitigate representation deviation. To address this, we propose Spatial-Spectral Visual Anchor Learning (SSVAL). The core of SSVAL is Visual Anchor Prompt Injection (VAPI), which introduces prompts that absorb rich knowledge from external VFMs during training, enabling them to serve as stable visual anchors that mitigate representation deviation during inference. Additionally, we incorporate auxiliary spatial and frequency-domain representation alignment losses to provide complementary vision-specific supervision at intermediate LLM layers. Extensive experiments demonstrate that SSVAL significantly outperforms existing methods. Code are available on our \href{https://msls38.github.io/SSVAL/}{project page}.

Vision-Language Models (VLMs) face a critical computational bottleneck when processing high-resolution imagery due to the $O(N^2)$ memory complexity of Softmax Multi-Head Attention (MHA). While substituting MHA with independent Multi-Layer Perceptrons (MLPs) achieves $O(N)$ scaling, it strips the architecture of spatial sequence routing, severely degrading global scene understanding and object permanence. In this paper, we propose the Linear Multi-Timescale Retention (LIA-MTR) module, a memory-efficient cross-modal bridge. By integrating an ELU-based positive feature mapping with adaptive write-gating and log-linearly distributed recurrent decays, LIA-MTR mathematically compresses continuous visual sequences into bounded memory states. Theoretical analysis proves the architecture operates with strict $O(N)$ sequence-interaction complexity. Empirically, synthetic retrieval evaluations demonstrate that LIA-MTR flawlessly routes context across 16,000 tokens, eliminating the "Lost in the Middle" degradation typical of naive linear attention. Hardware benchmarking reveals infinite-context scaling capabilities, natively processing 262,144 visual patches within an 11.2 GB VRAM footprint, whereas standard MHA suffers out-of-memory failure at 16,384 patches. Furthermore, following instruction tuning on 665K conversational samples, LIA-MTR significantly outperforms an industry-standard MLP baseline on the MME benchmark (71.00% vs. 68.11%), driven by a 10% absolute improvement in object permanence and superior global semantic extraction. This work establishes a mathematically rigorous, computationally flat foundation for infinite-context Vision-Language integration.

Language model benchmarking is a difficult task. Outcome reasoning alone does not test the model's conceptualization of language and popular open-source benchmarks are quickly saturated or ingested as training data. It is important to test the model's output, but augmenting these tests by characterizing semantic structure gives more insight to how models relate abstract concepts. However, the high dimensional embedding spaces are not easy to interpret. This work demonstrates how topological methods can be used to rigorously compare these spaces to low dimensional and interpretable baselines like ontologies and curated knowledge graphs. These multi-modal alignment tests make it possible to track model adaptations and test phrase understanding across multiple languages.

Interaction between users and LLM agents is increasingly multimodal: conversations interleave text with images, and a later question may target either. Yet most agent memories are designed around text, and even the few that support multimodal conversations still fail on vision-related questions. We trace this failure to an assumption behind the similarity search they rely on: in the index space, a query lies close to the relevant evidence that answers it. In multimodal settings, two gaps break it. By the modality gap, a query lies closer to memory content of its own modality than to evidence in another, even in a trained joint embedding space. By the similarity-relevance gap, the content most similar to a query is often not the evidence that answers it, most acutely when a query carries both text and image and its evidence resembles neither part alone. We present V-Mem, a multimodal agentic memory system that routes retrieval by the modality of the query and that of the target evidence, both recognized from the query alone. To cross the modality gap, V-Mem organizes the conversation into rounds and returns the target-modality content from the same round as the match, without comparing across modalities. To close the similarity-relevance gap, it searches with an LLM-generated anchor that sits closer to the relevant evidence than the query does: a hypothetical caption for a text-only query seeking an image, and an enriched search anchor, the query text plus relevant keywords extracted from the query image, when the evidence is reachable only by combining the two. On Mem-Gallery, V-Mem reaches an LLM-judge score of 0.82 versus 0.56 for the second best, with the largest margin on questions carrying an image (0.87, no baseline above 0.47); on LoCoMo it scores 0.69 versus 0.58.

Unified multimodal object tracking has achieved remarkable robustness by leveraging complementary sensor data (e.g., RGB, Thermal, Depth), yet the heavy computational burden of state-of-the-art models hinders their deployment on resource-constrained edge devices. In this work, we identify the prediction head as a critical but often overlooked efficiency bottleneck. By strategically streamlining the decoder architecture, we unlock the potential for real-time inference but simultaneously introduce a capacity gap between the lightweight student and the heavy teacher. To resolve this, we conduct a systematic analysis of 17 distillation strategies and introduce a Dual-Alignment Distillation framework. Our key insight is that effective compression requires decoupling knowledge transfer into two complementary streams: (1) Spatial Representation Alignment, which employs feature distillation to sharpen the student's spatial focus on foreground targets ("Where to track"); and (2) Semantic Distribution Alignment, which utilizes logit-based distillation to align decision boundaries and transfer discriminative dark knowledge ("What to track"). Extensive experiments across five benchmarks demonstrate that our approach significantly outperforms complex state-of-the-art methods. Notably, our distilled model achieves 91.5% MPR on RGBT234 and operates at 54 FPS on a single RTX 4090, representing a 5x speedup over the teacher model while maintaining superior accuracy.

Multimodal clinical models are usually judged on accuracy with every modality present, but deployment removes modalities; an echocardiogram is often unavailable where an ECG is routine. Two questions then matter beyond the size of the accuracy loss: which modality was responsible, and whether the model fails loudly or silently once that modality is dropped. The distinction is per-example and modality-level, and is separate from post-hoc feature attribution (e.g. SHAP). Models are replaced often; the evaluation that answers these questions is reused. We present a model-agnostic modality-failure framework: given N modality embeddings, any mask-aware probe, and labels, it returns a per-example failure taxonomy, a per-modality complementarity matrix that attributes error to modalities, and a loud-vs-silent dropout profile separating monitorable failures from those that pass unflagged far from the decision boundary, using only deployment-observable signals. We release it as a small, unit-tested harness and validate it against planted ground truth. Across seeds it recovers that planted modality dominance and complementary subset, reports per-modality loud-vs-silent rates, and scales to a three-modality complementarity matrix; because the planted structure is known by construction, this validates recovery of per-example attribution rather than clinical performance. We then instantiate the framework on frozen EchoJEPA and HuBERT-ECG embeddings for LVEF and the EF <= 40% HFrEF gate over a paired MIMIC-IV cohort, where on the held-out test split (n = 245) dropping echo nearly doubles error. The narrow echo-to-ECG overlap that bounds cohort size is itself a deployment finding for cardiac foundation models. All of our work can be found at https://github.com/criticaldata/PRIMED-AI.

Agents are increasingly expected to act not only as task executors, but also as decision-makers on behalf of human users. This shift requires agents to accumulate evidence over long horizons, interpret implicit user preferences, and compare multiple candidates under partial observations. In this work, we propose DunphyBench, a new benchmark for evaluating agents on long-horizon human-centered embodied decision-making, where the agent must navigate through multiple embodied housing environments and make decisions that align with multi-dimensional human preferences. Unlike standard embodied reasoning tasks that often focus on procedural planning or immediate goal completion, our setting requires agents to integrate multimodal, multi-source input into coherent knowledge that supports complex reasoning across long horizon. The evaluation results reveal that there is a substantial gap between current agents and human performance. Furthermore, our diagnosis of state-of-the-art VLM-driven agents reveals that memory management is one of the bottlenecks, where raw multimodal history introduces noise that hinders decision quality. Motivated by this finding, we design MeMento, a preference-conditioned multimodal memory compressor that selectively compresses decision-relevant information from long-horizon history based on user preferences with a fixed set of memory tokens. Experiments show that MeMento helps VLM-driven agents improve accuracy by 7.18%, while reducing memory usage by 85.38% compared to the strongest baseline.

The rapid dissemination of multimodal content has intensified the spread of fabricated news, presenting a substantial threat to social integrity. A formidable challenge for current detection systems is identifying misinformation related to novel events in zero-shot scenarios. Prevailing zero-shot methods typically assess news items in isolation via semantic matching, a strategy that fails to recognize the recycled disinformation tactics from past campaigns and lacks the sophisticated reasoning needed to identify subtle, cross-modal discrepancies. To surmount these deficiencies, we introduce \textbf{MRAFnd}, a novel \underline{\textbf{M}}ultimodal \underline{\textbf{R}}etrieval-\underline{\textbf{A}}ugmented Framework for Zero-Shot \underline{\textbf{F}}ake \underline{\textbf{N}}ews \underline{\textbf{D}}etection. MRAFnd emulates a collaborative team of analysts to verify news veracity. The framework initiates with \textbf{Multimodal Similarity-based News Retrieval} to assemble a corpus of contextually analogous articles from an unlabeled reference database. Subsequently, during the \textbf{Bifurcated Evidential Reasoning} stage, agents perform a dual-directional analysis to extract critical patterns from the retrieved evidence. Finally, a \textbf{Multi-Agent Collaborative Debate}, involving Analyst and Arbiter agents, engages in a structured discourse to arrive at a definitive and robust conclusion. Comprehensive experiments on three benchmark datasets reveal that MRAFnd markedly surpasses state-of-the-art baselines, achieving an accuracy gain of up to 2.35% on the demanding Weibo-21 dataset.

Recent studies develop pixel-level multimodal large language models (MLLMs) that support both Region Segmentation and Region Understanding, extending multimodal interaction from whole images to specific objects and regions. However, these methods face two fundamental challenges. First, the scarcity of high-quality mask--text pairs leaves abundant mask annotations without corresponding language supervision. Second, discrepancies in supervision formats and learning-signal densities induce optimization interference between Region Segmentation and Region Understanding. To address these challenges, we propose PixVL, a self-supervised post-training framework that introduces a unified Mask--Text Consistency Cycle, enabling pixel-level MLLMs to generate and self-verify regional descriptions and learn from unlabeled data. We found that direct cycle based solely on geometric reconstruction is unreliable because re-segmentation IoU does not faithfully reflect the semantic quality and referring sufficiency. PixVL therefore introduces confuser-aware semantic verification, which uses the model's confidence when it correctly chooses the target among highly similar candidate regions, and assigns zero reward to an incorrect choice. Meanwhile, PixVL performs cross-view verification using temporally separated video frames or geometrically transformed image views, preventing cyclic learning from collapsing to positional and shape shortcuts. Finally, a quality-coupled bidirectional learning strategy uses the highest-reward description to guide Text-to-Mask learning. This strategy transforms Region Understanding and Region Segmentation from competing tasks into mutual generators and verifiers. Experiments demonstrate that PixVL improves both region understanding task and segmentation task.

Multimodal large language models (MLLMs) increasingly rely on long chain-of-thought reasoning for complex tasks. However, as reasoning sequences lengthen, models may gradually rely less on visual evidence and more on accumulated textual context, leading to visual forgetting. Existing approaches do not directly constrain how visual evidence is used and maintained along the original reasoning trajectory, leaving long-context visual forgetting insufficiently addressed. To address this issue, we propose Remember-R1, a reinforcement learning framework that mitigates long-context visual forgetting by applying process-level supervision directly on the original reasoning trajectory. Specifically, Remember-R1 introduces rewards that encourage broader coverage of matched visual keywords, stronger persistence of visual dependence in later reasoning steps, and greater focus on question-relevant image regions. Experiments across multiple model scales and diverse multimodal benchmarks demonstrate that Remember-R1 consistently improves reasoning performance. Additional analyses further show that it slows the decline of visual attention during generation, supporting its effectiveness in mitigating long-context visual forgetting.

State-of-the-art RGB-Event detectors improve the detection of small, fast-moving objects by combining complementary features from RGB and Event data, yet they typically fuse the two modalities into a unified representation for both localization and classification. Such a task-symmetric design is inconsistent with the intuition that the two modalities should play different roles according to their task-specific strengths. To examine this issue, we conduct a modality-specific evaluation and find that the relative advantage of the two modalities reverses across tasks: Event data are substantially more effective for class-agnostic localization, whereas RGB data provide stronger category evidence within localized target regions. Motivated by this task-dependent asymmetry, we propose an Asymmetric Event-RGB Object Detection Transformer (AERODet). During class-agnostic localization, Scale-wise Uncertainty-aware Reliability Estimation (SURE) calculates the relative reliability of the two modalities from their objectness response heatmaps and accordingly calibrates their contributions when the decoder aggregates multimodal features. Once the candidate boxes are obtained, Task-Decoupled Semantic Refinement (TDSR) decouples classification from localization and uses RGB RoI features for fine-grained classification. Extensive experiments on FRED and NeRDD demonstrate that AERODet achieves state-of-the-art performance. In particular, it surpasses the strongest RGB-Event baseline by 10.7 mAP points on the FRED challenging split.

On-policy distillation (OPD) samples trajectories from the current student policy and minimizes token-level divergence between student and teacher next-token distributions at prefixes along those trajectories. This aligns the distillation states with the student's own generation distribution. However, it still assumes that the complete teacher distribution is an appropriate target across student capacities. In vision--language reasoning, teacher corrections can depend on visual distinctions that a compact student cannot represent. Our target-scaling study shows that, as the target approaches the complete teacher distribution, the student realizes less of the prescribed shift and obtains worse downstream performance. We therefore propose \emph{Fisher-Projected On-Policy Distillation} (FP-OPD), which distills only locally realizable teacher corrections. FP-OPD uses continuous visual perturbations to estimate the student's local visual tangent space and projects the centered teacher--student log-probability gap onto this space under the student's Fisher metric. The resulting capacity-aware target is optimized with full-vocabulary reverse KL on student trajectories, retaining the standard OPD framework. In 8B-to-2B distillation, FP-OPD improves all seven evaluated multimodal benchmarks. It raises the average score by 2.77 points over the pretrained student and by 1.60 points over standard OPD. These results demonstrate that locally realizable teacher corrections provide a more effective target for distilling compact vision--language models.

Vision Language Models (VLMs) have demonstrated exceptional performance across various tasks. However, they have not yet been thoroughly evaluated on more complex tasks. The Persuasion Model, conceived by Aristotle, resembles a triangle shape, which highlights its inherent challenges related to personal biases. To assess the progress of VLMs on these complex tasks, we use the ImageArg datasets, focusing on the Logos, Ethos, and Pathos detection tasks. Our findings indicate that models from the Qwen family achieve improved F1 scores, with Qwen3 performing exceptionally well on the Logos and Pathos tasks, while Qwen2 exhibits competitive performance on the more complex Ethos detection task. We release the code to foster research in this direction.

Multimodal large language models (MLLMs) exhibit strong vision--language capabilities but may also memorize and disclose sensitive information. Machine unlearning seeks to remove designated knowledge without retraining from scratch while preserving general utility. Existing privacy-oriented benchmarks primarily adopt profile-level deletion, whereas practical requests are often finer grained: a model should forget a specified attribute while retaining non-sensitive information about the same identity. We therefore introduce attribute-level MLLM unlearning as a finer-grained task and construct a benchmark spanning long-text, numeric, and short-text targets, multiple forget ratios, and diverse question types. Our evaluation reveals that target and retained attributes share identity-specific and visual evidence, making selective forgetting susceptible to residual leakage or collateral degradation; accordingly, existing methods exhibit unstable forgetting--retention trade-offs in this setting. To address this challenge, we propose Causal Localization and Retain-Aware Projection (CLRP), a lightweight training-free framework. CLRP uses activation patching to identify the layer that causally mediates target-attribute disclosure, then applies a retain-aware projection that removes the target-attribute subspace while preserving same-identity evidence. Experiments across multiple widely used MLLMs with distinct architectures and parameter scales demonstrate the effectiveness of CLRP.

Large vision--language models (VLMs) can reason step by step about complex visual scenes, but this open-ended, autoregressive chain-of-thought (CoT) approach is poorly suited to safety-critical, rule-governed settings such as industrial surveillance, where decisions must be bounded, deterministic, and low-latency. Because CoT inference cost scales jointly with reasoning length and the number of concurrent streams, it creates a throughput bottleneck that precludes the real-time, multistream monitoring required for industrial accountability. Here we present MonitorVLM-v2, a deployment-oriented framework that recasts VLM-based safety assessment as probabilistic inference over a finite regulatory decision space, compressing multimodal reasoning into single-step rule-ID predictions and reducing decoding from a variable-length sequence to a single token. We introduce symbolic policy optimization (SymPO), a novel contrastive policy optimization algorithm that sharpens decision boundaries within this finite symbolic space, together with an entropy-driven triage mechanism that routes uncertain predictions to human reviewers for expert confirmation. In a four-month prospective deployment across 10 concurrent camera feeds in an operational underground mining facility, MonitorVLM-v2 achieved a 19.45-fold increase in inference speed and identified 2.78 times as many confirmed violations as the site's routine manual inspection workflow, demonstrating the practical value of compressed symbolic decision-making for real-time, auditable industrial monitoring.

Remote-sensing multimodal large language models (MLLMs) often assert facts that imagery cannot establish, such as a facility's identity or function. Coordinate-keyed geographic retrieval can supply this missing knowledge, improving fMoW land-use accuracy by 12.06--17.19 points across three open MLLMs. However, retrieved records can also contradict visible evidence, and we find that models frequently follow the records even when the image is decisive. We argue that source trust should therefore depend on \emph{cross-modal verifiability}: geographic records are most useful for attributes the image cannot verify and most dangerous when they dispute visually verifiable attributes. We introduce GeoArbiter, a training-free pipeline that operationalizes this principle by injecting only image-unverifiable geographic facts. Unlike arbitration prompts, which leak across attribute types and bias yes/no responses, content-level filtering preserves 84.69--87.15% of the full-retrieval accuracy gain, reduces claim-level hallucination by 9.58--26.34% under a source-blinded judge, and improves robustness to conflicting records across all three models. These results identify verifiability-guided content selection as a simple, effective mechanism for grounding remote-sensing MLLMs in fallible geographic knowledge.

Traditional reaction yield prediction is constrained by 1D quantum descriptors that lack explicit spatial information. To address this gap, a dual-modal Vision Cross-Attention architecture is proposed, fusing tabular physical-organic data with 2D molecular topologies. Notably, it is demonstrated that a generic computer vision backbone processing simple 2D skeletal structures independently outperforms purely quantum-based baselines. By synergizing both modalities, superior predictive accuracy compared to traditional methodologies is achieved by the optimal cross-attention framework (Test RMSE = 5.27%). Through mechanistic probing, active, descriptor-guided spatial querying is observed, effectively offloading macroscopic steric identification to the visual pathway. Furthermore, a dynamic chemical hierarchy is learned by the network to heavily prioritize critical steric bottlenecks, such as the aryl halide. Concurrently, residual skip connections are utilized to protect non-spatial electronic parameters from destructive attenuation during fusion. Collectively, a scalable and highly interpretable blueprint is provided for augmenting physical chemistry with deep visual learning.

Multimodal-attributed graphs (MAGs), where nodes carry heterogeneous semantic content across multiple modalities while edges encode relational dependencies, have been widely adopted across diverse domains. Federated multimodal graph learning (FMGL) extends federated graph learning (FGL) to MAGs, enabling collaborative optimization across decentralized MAGs without exposing raw data. However, naively applying existing FGL methods to FMGL is insufficient, as they fail to navigate the multifaceted heterogeneity inherent in decentralized MAGs, including task heterogeneity across diverse client objectives, modality heterogeneity from discrepant modality quality and semantic domains, and topology heterogeneity arising from divergent topological patterns with low cross-modality correlation. To address these challenges, we propose Federated multimodal graph learning with Topology-aware Cross-modal Routing (FedTCR), the first systematic algorithm designed for FMGL. To handle task heterogeneity, FedTCR employs a two-stage paradigm that comprises federated task-agnostic pre-training followed by isolated task-oriented fine-tuning. To jointly address modality and topology heterogeneity, FedTCR introduces a topology-aware cross-modal routing mechanism. Concretely, each client distills modality-specific knowledge into compact prototypes via topology-aware importance-weighted aggregation informed by graph structure; the server then evaluates cross-client cross-modal relationships among these structure-informed prototypes and routes informative ones as contrastive references, driving a tri-level cross-modal contrastive learning scheme that jointly aligns cross-client modalities while preserving discrimination. Experiments across 7 domains demonstrate that FedTCR outperforms state-of-the-art baselines on both graph-centric and modality-centric tasks.

Vision-language models (VLMs) process image patches and text tokens in a shared residual stream, but the local geometry through which the two modalities interact remains poorly understood. Most interpretability methods identify global linear directions, which may miss representations that are globally high-dimensional but locally low-dimensional. We introduce LENS (Local Explanation of Neighborhood Subspaces), a method that decomposes VLM activations into local low-rank Gaussian neighborhoods using a Mixture of Factor Analyzers. Applied to LLaVA-1.5-7B and Qwen3-VL-8B, LENS reveals distinct depth-dependent fusion trajectories consistent with each model's fusion mechanism: LLaVA progressively mixes modalities at later layers, whereas Qwen3-VL mixes them early, partially re-segregates them, and recombines them near the output. An automated multimodal labeling pipeline assigns concise semantic descriptions to these neighborhoods. Interpolating activations toward neighborhood centroids causally redirects generation within and across modalities and outperforms difference-in-means and VL-SAE in most evaluated conditions; in one LLaVA vision-to-vision setting, MFA achieves 5.7 times the VL-SAE score. Human evaluation finds MFA steering competitive with prompting and substantially stronger than the other intervention baselines. Finally, the MFA coefficient space improves Qwen3-VL image-to-rendered-text retrieval at the deepest evaluated layer from 14.9% to 48.6% R@1. Ablations show that the reported fusion trajectories are stable across component counts, local ranks, and modality-purity thresholds. These results support local geometric neighborhoods as useful interpretable and causal units for analyzing cross-modal representations in the evaluated VLMs.

Infrared-visible image fusion under realistic degradation scenarios is a challenging task, as degradations not only cause a loss of reliable modality-specific information in observed images but also hinder the fusion process. Recent studies indicate that text can provide prior information about degradation characteristics, complementing the limited evidence available from corrupted input images and facilitating fusion. However, existing methods typically inject fixed global text representations into visual features, making it difficult for textual guidance to adapt to spatially varying degradations, local structures, and thermal saliency. To this end, we propose TGFusion, a text-guided latent-space flow matching framework that unifies degradation suppression and cross-modal fusion. TGFusion encodes task, degradation, and generation cues into structured prompts. To fully exploit these priors, we design a Prompt-conditioned Multi-stream Joint Flow Transformer that represents text as an independent semantic stream alongside fusion, visible, and infrared streams. Joint attention enables token-level bidirectional interaction and layer-wise updating among semantic and visual representations, allowing degradation semantics to dynamically guide reliable information selection and fusion latent generation. Extensive experiments on public benchmarks and complex degradation scenarios demonstrate that TGFusion achieves superior or competitive performance in perceptual quality, image naturalness, structural-detail preservation, and infrared-saliency retention, while remaining robust across diverse single and compound degradations.

Human language is highly polysemous. Many common words (e.g., 'bank' or 'palm') carry several distinct meanings that shape what humans communicate and imagine. Large language models (LLMs) have been shown to understand this multiplicity of meaning, but much less is known about how polysemy surfaces in other modalities such as images. We study this across 17 text-to-image and 15 text-generation models by giving each a polysemous word with no context to fix its meaning and measuring which senses are produced over many samples. We find a clear multimodal gap, where within every model family, generated images settle on far fewer senses than generated sentences (normalized entropy 0.10 vs. 0.25), and both are far less varied than what people imagine for the same words (normalized entropy 0.47). However, when we instead ask a model to list how often it would generate outputs corresponding to each possible meaning of a word, it predicts distributions that are more diverse than the actual space of outputs. These results reveal a multimodal gap in how foundation models express meaning, and how their understanding may not transfer faithfully nor equally across modalities.

Traditional static assessments rely on a subtractive, deficit-based grading model that often penalizes ambition and obscures diagnostic feedback. Conversely, traditional face-to-face oral examinations introduce severe construct-irrelevant variance by exacerbating performative anxiety and the sociological power imbalances inherent to academic hierarchies. This paper presents the theoretical foundation for the "Socratic Test," an automated, computer-mediated conversational assessment. By integrating Dynamic Assessment principles, multimodal workspaces, Bloom's Taxonomy for real-time proctoring, and the SOLO Taxonomy for structural evaluation, the Socratic Test actively maps a student's cognitive boundaries. This paper formalizes the use of graduated scaffolding to quantify the Zone of Proximal Development (ZPD) and details a non-compensatory, additive grading architecture that prioritizes mastery over penalty and human-AI alignment to ensure unprecedented measurement reliability.

Multimodal video generation aims to generate and edit videos conditioned on arbitrary combinations of text, images, and videos within a single model, allowing diverse tasks to share complementary data and generative priors. Unifying these tasks requires multimodal understanding of diverse conditions, which is typically provided by a pretrained vision-language model (VLM). A key challenge is how to connect the VLM's hierarchical multimodal representations with a pretrained video diffusion transformer (DiT). Existing methods either inject features from only the final or a few manually selected VLM layers, or jointly train architecture-matched understanding and generation streams, making it difficult to reuse heterogeneous pretrained backbones. We introduce MoRoute, a unified multimodal video generation framework that formulates a frozen VLM and a pretrained video DiT with different architectures as heterogeneous experts connected through dynamic layer routing. For each input, a lightweight block-wise router enables every DiT block to select the VLM layer most relevant to its generation stage, thereby learning an adaptive correspondence between multimodal understanding and video synthesis. MoRoute further incorporates reference images and source videos directly into the DiT token sequence through unified in-context conditioning, preserving fine-grained visual details across diverse generation and editing tasks. Experiments on IntelligentVBench, OpenVE-Bench, and RefVIE-Bench show that MoRoute consistently surpasses the best competing method on each benchmark, improving the average score by 0.15, 0.18, and 0.34 on a 1-5 scale, respectively.

Modern recommender systems in food delivery increasingly leverage multimodal signals, including images, text, and user interaction histories, to enhance user experience, yet effective fusion of these heterogeneous modalities remains challenging, hindering both the joint modeling of multimodal signals and adaptation to evolving user intent. In mainstream two-stage approaches, the separation between content-semantic pretraining of image-text encoders and behavior-driven ranking models limits alignment between semantic understanding and user behavior patterns. To address these issues, we present GALA, a three-stage pipeline whose core innovation lies in an intermediate "generative RL alignment" stage that constructs multimodal pretraining data from user behavior and refines it via conversion-based rewards, effectively bridging the pretraining-fine-tuning gap to align with downstream objectives. GALA comprises three stages: first, behavior-aware triplet pretraining on query-image-text pairs from search logs to early capture user intent and content preferences; second, a novel intermediate stage that refines multimodal embeddings through reward-driven optimization (GRPO) to dynamically align them with user behavior and bridge the pretraining-fine-tuning gap; and finally, integration of multimodal and ID embeddings via adaptive gating with a hybrid loss, preserving multimodal contributions under long-term ID-dominant training. GALA has been deployed in the production environment at Taobao Shangou, serving over 200 million daily active users. Compared with state-of-the-art (SOTA) methods, it delivers consistent offline gains of +0.12/+0.20 AUC along with better PCOC metrics. Large-scale online A/B tests further report a 0.55 percent increase in order volume, confirming GALA's effectiveness at industrial scale and its robustness across diverse demand patterns.

Audio-visual speech recognition (AVSR) relies on effective fusion of audio and visual modalities, yet existing approaches treat cross-modal interaction as a single-step operation without structured iterative refinement. We present DoubleHelix, a multimodal fusion framework that reformulates fusion as an iterative cross-modal interaction process with adaptive degradation-aware enhancement. The framework comprises three components including ReverseParallelHelix for multi-turn structured interaction with learned alignment constraints, QualitySensor for learning degradation-aware gating signals, and HelixReplication for consistency-guided conditional feature enhancement. Experiments on LRS3 demonstrate that DoubleHelix achieves 0.68% WER on clean audio, outperforming previous best results by 5.6% relative improvement under matched backbone settings. Comprehensive ablation studies validate each component contribution, including targeted analysis of design choices such as asymmetric pathway weighting. The framework shows improved robustness under evaluated babble-noise conditions, achieving 11.6% WER at SNR -5dB.

Training-free acceleration makes diffusion-based multimodal large language models (dMLLMs) more deployable, but it may silently change generated content. We study this serving-time consistency problem on 300 real images, comparing Fast-dLLM outputs with the same model's unaccelerated outputs. Across the mild parallelism induced in our long-form setting (1.05--1.25 committed tokens per step), confidence-threshold tuning changes decoding behavior but not baseline agreement. State-refresh ablations and an image-swap intervention instead identify stale visual and generated-text states as contributors to drift. For the tested Fast-dLLM implementation, shortening the KV-cache refresh interval yields a monotonic speed--agreement frontier and near-exact agreement at a measured 1.3x speedup. The initial diagnosis also appears with dLLM-Cache and LaViDa, although dLLM-Cache recovers agreement only after both caches are tightened, which removes its speed advantage. Independent prompts and images reproduce the threshold-insensitivity and refresh recovery. A targeted audit finds genuine content substitution in half of 50 low-agreement pairs. In a separate blinded two-annotator evaluation, the pooled accelerated-minus-baseline factual-error difference is 0.00 (95% CI [-0.17,+0.17]); this sample detects no difference but does not establish factual equivalence. Finally, none of the tested adaptive or smoothed-refresh variants beats the fixed interval at matched compute. Our contribution is a paired diagnostic and an implementation-scoped consistency control, not an accuracy or safety guarantee.

While recent image editing models have made rapid progress, multi-reference editing remains challenging, particularly in maintaining visual consistency across references and ensuring overall visual harmony. Reinforcement learning has proven highly effective for text-to-image generation and single-image editing, but its extension to multi-reference editing is hindered by the absence of suitable reward models that capture multi-image relational constraints. Moreover, naively using multimodal large language models(MLLMs) as zero-shot evaluators faces a key tension between hallucination-prone long-form reasoning and the limited deductive power of short-form judgments. We address these issues with a Multi-dimensional Evaluation-Verification Reward(EVR). EVR decomposes evaluation into distinct visual criteria; for each criterion, an MLLM Evaluator generates multiple candidate hypotheses, and a Verifier grounds each claim in concrete visual evidence to accept or reject it, producing reliable and fine-grained reward signals. Together with a scalable data pipeline, our method enables RL fine-tuning of off-the-shelf editors without architectural changes. Extensive experiments show substantial gains over the base Qwen-Image-Edit, improving consistency and harmony to match or surpass NanoBanana.

Modern opaque AI models prize performance over interpretability, which makes testing difficult. However, formal statistical tests conducted on a model's embedding space can provide robust characterizations of semantic structure, concept separation, and knowledge graph alignment. Model developers would benefit from a model comparison technique that leverages human-curated knowledge structures to test alignment. The scale of the input space for even relatively simple tasks motivates the need for alignment checks that augment standard outcome reasoning. This work develops and demonstrates a topology-based multi-modal alignment test to make deployment, selection, and comparison of opaque models more interpretable. These methods also offer an intuitive connection to possibility theory and a unified decision theoretic framework from data to deployment.

Online shoppers increasingly turn to AI shopping assistants, using images and multi-turn dialogue to express and refine product needs that are difficult to articulate in text alone. However, existing benchmarks largely rely on text-only or synthetic requests, underrepresenting complex real-world shopping requirements jointly expressed through images and language. We introduce MMShopBench, the first real-log benchmark for multimodal, multi-turn shopping agents. Built from carefully cleaned and manually annotated shopping logs, MMShopBench provides ground-truth annotations of each request's purchase intent and mandatory product requirements. Agents must infer these requirements jointly from user images and multi-turn dialogue, retrieve candidate products through image and text search, and verify that each candidate satisfies all requirements using its product images and structured attributes. We evaluate representative open-source and proprietary models using an evidence-grounded multimodal protocol and construct a companion training set for fine-tuning an open-source model. To ensure reproducible experimentation, we build an offline shopping sandbox, where fine-tuning substantially narrows the performance gap between our open-source model and leading proprietary models, demonstrating the effectiveness of our training data.

Multimodal information can improve the accuracy of click-through rate (CTR) prediction and effectively alleviate item cold-start and long-tail problems. Recent studies commonly discretize pretrained multimodal embeddings into semantic identifiers (SIDs), allowing the model to learn task-specific semantic representations for recommendation. However, existing methods still provide limited gains due to two major limitations. First, codebook assignment fails to preserve semantic relevance and discards fine-grained continuous signals in the original embedding space. Second, the residual code paths are highly dependent on prefix codes, which limits the effective representational scalability of hierarchical identifiers. To address these issues, we propose PaletteID (PID), a prototype-based semantic identifier. Inspired by palette-based color composition, PID uses a compact set of representative prototype items as semantic anchors to bridge pretrained multimodal content space and recommendation models. Specifically, we first construct a prototype palette with Semantic Quality-Aware Determinantal Point Process (SQ-DPP), which jointly considers local content density and global semantic diversity. Then, for each target item, PID retrieves a sequence of semantically related prototypes and aggregates them into an informative PID representation, enabling rich and complementary semantic modeling. Extensive experiments on two public datasets demonstrate that PID consistently improves CTR prediction and yields larger gains for long-tail items. PID also produces more robust identifier assignments and provides more interpretable token semantics than existing residual SID methods.

Although Large Language Models (LLMs) have demonstrated promising safety performance, extending them to Multimodal Large Language Models (MLLMs) exposes a significant gap between expanded multimodal capabilities and existing safety mechanisms. Current defenses remain predominantly confined to specific modal settings, thereby limiting their robustness against broader cross-modal threats. To bridge this gap, we introduce SafeNexus, a cross-modal safety alignment framework that adopts a dedicated neuron-level intervention strategy. First, we formulate a neuron localization paradigm that identifies functionally specialized neurons by characterizing intermediate-layer activation patterns and quantifying their functional salience through importance scoring. Building upon this paradigm, we exploit contrastive data to identify modality-bound safety neurons (BS-Neurons), and validate their role in regulating safety behavior within each modality via targeted suppression. Further cross-modal analysis defines modality-universal safety neurons (US-Neurons) as the shared subset of BS-Neurons identified across individual modalities, serving as the core for defending against harmful cross-modal attacks. We observe that suppressing these neurons substantially degrades safety performance across modalities, while leaving overall utility largely unaffected. Building on these insights, we propose two safety alignment strategies: activation-level safety amplifier and safety neuron calibrator. The proposed strategies enhance model safety through two distinct routes: the former amplifies the activation magnitudes of US-Neurons, while the latter selectively calibrates them via targeted fine-tuning. Extensive experiments demonstrate that our method outperforms prevailing state-of-the-art approaches on safety benchmarks spanning diverse modality combinations, while effectively preserving utility.

Universal multimodal embedding (UME) maps heterogeneous multimodal inputs into a shared embedding space. Existing UME models either form embeddings through single forward encoding or add computation through explicit rationale tokens and latent autoregressive states. Although token expansion can improve complex matching, serial generation increases retrieval latency and makes the final embedding depend on generated intermediate states. This raises a different question: can useful computation be expanded along model depth while keeping the token workspace fixed? We analyze positive-negative similarity separation at every layer of independently trained UME models and observe a shared progression: early layers contextualize multimodal inputs, a contiguous middle-to-late stage forms retrieval-discriminative features, and the final layers map them into the embedding space. Based on this finding, we propose ReLoop-UME, which executes the early layers once, recurrently reuses a parameter-shared retrieval-forming block, and applies the final mapping layers after the last loop. Learnable Retrieval Registers provide persistent retrieval-specific states that accumulate and exchange evidence across loops, with the final register serving as the embedding readout. On MMEB-V2 and MRMR, ReLoop-UME consistently improves retrieval across different backbones while running 44.9x faster than UME-R1 and 1.5x faster than PLUME.

Multimodal federated graph learning enables clients to collaboratively train graph models over structural, textual, and visual signals without sharing private local data. However, the presence of heterogeneous multimodal content also makes unlearning requests more frequent and fine-grained: users may delete accounts or interactions, remove a particular image or text while retaining the associated entity, or revoke the learned correspondence between retained modalities or graph attributes. Existing federated graph unlearning mainly handles entity/relation or client removal and cannot directly satisfy these multimodal requests. They introduce three challenges: removing only the requested information without damaging retained content, preventing the target from being recovered through remaining modalities or graph neighborhoods, and stopping related traces on other clients from re-entering the global model after aggregation. To address them, we propose \textsc{\textbf{MMFGU}}, a multimodal federated graph unlearning framework built around target-specific representation decoupling. \textsc{MMFGU} maps heterogeneous requests into unified target carriers, decouples requested representations while anchoring retained semantics, exposes and repairs propagated residuals with lightweight probes, and selectively purges affected clients through compact prototype and response signals. Experiments show that \textsc{MMFGU} effectively removes requested information, preserves retained graph utility, and achieves a $\boldsymbol{41.5\times}$ speedup over full retraining.