Multimodal Embedding

Multimodal embedding spaces in models like CLIP enable powerful capabilities such as semantic similarity retrieval and cross-modal zero-shot classification. These embeddings compress high-level semantics into a single vector, which comes at the cost of primarily expressing a dominant semantics like main object while suppressing other important attributes such as camera angle or color tone. We propose a text-conditioned transformation of visual embeddings that makes such attributes explicitly accessible. Given a natural language description of an attribute category (e.g., "color" or "art style"), a network generates an affine transformation that emphasizes the specified attribute. Conditioning on text enables it to learn many attributes simultaneously, accessing them at inference time through an intuitive interface. The network is trained to align transformed embeddings with the frozen latent space, enabling retrieval using existing large-scale embeddings without any re-encoding. When applied to a full set, the same mechanism transforms the latent space for attribute disentanglement tasks such as multi-clustering. By operating directly in latent space, our method provides a unified and efficient framework for controlling embedding spaces, demonstrating state-of-the-art performance across both attribute-based retrieval and multi-attribute organization tasks with near-zero inference cost. Project page: https://joefioresi718.github.io/ControlEmbed_webpage/

The renewed global focus on lunar exploration, driven by the prospect of in-situ resource utilization and a sustained human presence on the Moon, has created growing demand for accurate, large-scale characterization of the lunar surface. Although vast quantities of orbital remote-sensing data have been collected, scientific analysis and resource mapping remain fragmented by heterogeneous multiinstrument observations, sparse labels, and bespoke task-specific modelling workflows. Here we introduce LunarFM, a multimodal foundation model that learns a general representation of the lunar surface from diverse orbital measurements. LunarFM assimilates observations from six instruments across three lunar missions, mapping 18 input channels to a shared embedding space. We demonstrate that this embedding space supports a diverse range of downstream applications, including similarity search, few-shot resource mapping, mineral abundance regression, and geological unit classification, enabling efficient scientific investigation and resource-oriented analysis. We provide a machine-learning-ready dataset of co-registered multimodal observations spanning latitudes from 70°S to 70°N, a pretrained multimodal masked autoencoder, and a companion embedding dataset providing a joint 768-dimensional representation of lunar surface properties. All code and data are available at https://lunarfm.trillium.tech/

A single embedding space that covers text, images, video, and audio lets one index serve every query a user can pose. Embedding models built on vision-language backbones now lead text/image/video retrieval benchmarks but lack audio entirely, while audio-text retrieval is led by specialist systems that serve no other modality. We present the Fusion Embedding family, which adds audio to a frozen vision-language embedding base whose parameters are never updated: generation 1 (fusion-embedding-1) trains only a 16.4M-parameter connector between a frozen audio tower and the frozen base, and generation 2 (fusion-embedding-2) adds modality-gated deep adapters (44.2M parameters) whose branch never executes on text, image, or video inputs: their outputs are bit-for-bit those of the released base, verified after every training run. Because the base already binds text, images, and video, aligning audio to text alone makes audio-image retrieval emerge, with zero paired audio-visual training data. Alongside the recipe we map its design space with controlled negative results (rewriting training captions with an LLM, substituting a leaderboard-stronger audio tower, and widening the connector each reduce retrieval) and with training-protocol findings that we expect to transfer to any frozen decoder-LM embedding backbone. Both generations train in hours on a single GPU. Weights, code, and the evaluation harness are openly released.

The Contrastive Olfaction-Language-Image Pre-training 2 (COLIP-2) model is a multimodal embeddings space that places olfaction as a first-class citizen among vision and language. Molecular structure, gas-sensor readings, odor-descriptor language, and images are all trained into a single shared representation space, so that a robot can localize a detected aroma to objects in a scene probabilistically. No ImageNet-scale datasets of paired image-scent examples exists which warrants the need for their collection. Our intent with the release of COLIP-2 is to demonstrate the limit of what can be built for robotics with open-sourced olfactory data in order to ground the argument for why new methodologies and datasets are necessary in order to enable advanced olfactory-oriented perception capabilities. We enumerate results from internal testing of the COLIP-2 architecture and make necessary optimizations to run the model at the edge for real-time robotics applications. While developed with robotics in mind, the design of COLIP-2 has been influenced by experts across many disciplines of science in academia and industry, and we hope that the model can be useful in any multimodal domain requiring olfactory intelligence.

Text-motion representation learning has advanced rapidly, with growing interest in multi person interactions for animation, AR/VR, and embodied AI. These settings require representations that align language with both individual actor dynamics and the relationships between actors. We introduce the Multimodal Interactive Motion Encoder (MIME), which, to our knowledge, represents the first dedicated multimodal encoder designed specifically for two person interactive motion. MIME captures individual and shared structure using stream based co-attention with explicit interaction features and curriculum based contrastive training. On Inter-X text-motion retrieval, MIME consistently outperforms early and late fusion baselines across gallery sizes, achieving a 12.8% relative improvement in text-to-motion R@1 at a 2,000-sample gallery. We further evaluate MIME as a frozen auxiliary prior within TIMotion and InterMask on the unseen InterHuman dataset. MIME improves semantic alignment metrics while maintaining comparable FID in TIMotion. These results show that interaction aware multimodal encoding improves multi person motion retrieval and transfers across datasets to support downstream motion generation.

Spatially resolved omics studies increasingly combine transcriptomic and epigenomic assays, yet downstream analysis is often still performed using single-modality pipelines. We present LATTICE (Latent Alignment of Tissue-level and Transcriptomic Information for Cross-modal Embedding), a graph-based self-supervised framework that learns spot-level representations from harmonized multimodal features. LATTICE integrates five aligned modality blocks per Visium spot: Visium RNA, scMultiome RNA, scMultiome ATAC, spatial ATAC, and spatial CUT&Tag. These modalities capture spatial transcriptomic measurements, single-cell inferred regulatory activity, and in situ chromatin and histone states within a unified lattice representation. LATTICE constructs a spatial neighborhood graph and trains a TransformerConv encoder using masked reconstruction, cross-modal alignment, and spatial smoothness objectives. On a private 11-sample melanoma cohort from an anonymized clinical collaborator comprising 54{,}912 total spots, LATTICE demonstrated stable optimization behavior, reproducible embeddings across analysis seeds, and complete multimodal integration across all samples. Adding scMultiome RNA to Visium RNA alone substantially improved concordance with Space Ranger clusters across 11 runs (adjusted Rand index [ARI] +0.157, normalized mutual information [NMI] +0.143, and spatial contiguity +0.174). Additional modalities further improved spatial contiguity and multimodal utility score (MUS), although they sometimes reduced agreement with RNA-derived reference labels, likely because the learned embeddings captured chromatin and regulatory structure beyond transcriptomic similarity alone. These results position LATTICE as a practical and empirically grounded framework for multimodal spatial omics integration, while also highlighting the need for stronger supervision and broader external benchmarking.

Contrastive Language-Image Pretraining (CLIP) representations form a semantic embedding space governed by cosine similarity, reflecting an intrinsic hyperspherical geometry. However, existing probabilistic interpretations typically rely on Gaussian assumptions, which fail to capture this directional and multimodal structure. We propose a principled density model for the CLIP latent space based on Mixtures of von Mises-Fisher (MovMF) distributions defined on the unit hypersphere. Using the Expectation-Maximization (EM) algorithm, we efficiently learn a probabilistic model in which each mixture component corresponds to a coherent semantic concept. This formulation yields a closed-form likelihood naturally aligned with hyperspherical geometry, enabling accurate and interpretable density estimation. Empirically, our model significantly improves long-tailed and out-of-distribution detection and provides a natural semantic decomposition, representing each embedding as a sparse probabilistic combination of interpretable concepts. These results suggest that CLIP latent space is more faithfully characterized as a hyperspherical semantic mixture rather than an isotropic Gaussian, establishing a simple and geometrically consistent probabilistic framework for modeling and understanding multimodal representations. Project page is available at https://xiaoyuzhizi.github.io/movmf-clip/.

A robot must understand the state of its own body, but a camera sees only part of it. Force and contact leave almost no trace in a single frame, and raw vision features read force at $R^2$ at or below $0.10$ on every robot we test. We present Kepler-Encoder-v0.1, a robot-first multimodal encoder that treats robot state as a modality and fuses vision, proprioception, and force/torque into a single shared latent with a learned-query cross-attention layer, trained self-supervised by masked cross-modal prediction under the LeJEPA/SIGReg objective. At evaluation only vision enters, which poses a sharp question. Does fusing state into training make the vision-only latent carry anything the pixels do not already contain? On the RH20T corpus the answer is yes, precisely where the camera is weakest. On held-out scenes, the vision-only latent recovers end-effector state, and force in particular, significantly above both raw frozen-ViT features and a compute-matched vision-only control on every sensored robot, though absolute force recovery at a single timestep is modest; on motor state, which the camera largely sees, it is statistically tied with the strongest vision baselines, and it is the only feature whose latent geometry tracks state. A single embodiment-agnostic encoder covers four robots, and a data-matched control shows this breadth reflects embodiment diversity rather than data volume. The frozen latent is directly useful. Its own cross-modal prediction error is a training-free invalid-state monitor (AUROC $0.90$ on out-of-range states, $0.69$ on scene-swapped states), and a diffusion decoder (PixNerd) reconstructs the camera frame from the latent, confirming the spatial compression preserves world-state. This report validates the single-timestep case; native-rate temporal fusion is the next step.

Multimodal information is pivotal for e-commerce search ranking. Existing works leverage multimodal data typically by fine-tuning general Multimodal Large Language Models (MLLMs) via collaborative signals, subsequently integrating the derived representations into ranking models as item features. Despite their efficacy, these methods face two primary limitations: (1) they rely on a single collaborative signal for MLLM fine-tuning, failing to exploit the heterogeneous signals essential for multitask ranking; and (2) they treat multimodal representations as regular item features in ranking models, underutilizing their latent potential for user behavior modeling. To address these challenges, we propose the Multiplex Multimodal Representation Model (MMRM), a unified framework that aligns MLLMs with diverse collaborative signals. By employing a shared backbone with task-specific tokens and projection layers, MMRM simultaneously learns from multiple signals and generates comprehensive multiplex item representations in a single inference pass. Furthermore, we introduce a multiplex user representation strategy in ranking models, which derives task-specific user representations via search-based behavior sequence modeling leveraging multiplex item representations. Extensive experiments demonstrate MMRM's superior efficiency and effectiveness. Notably, MMRM has been successfully deployed in the JD e-commerce search engine, yielding significant performance gains for millions of daily users.

Few-shot multimodal classification commonly attaches a lightweight head, such as $k$-nearest neighbors, logistic regression, or a linear SVM, to a frozen pretrained encoder. Although computationally efficient, these heads can produce poorly calibrated confidence scores, limiting their reliability in calibration-sensitive applications. We evaluate TabPFN as a plug-and-play, zero-gradient classification head for frozen image, text, and audio encoders. Across 22{,}820 evaluation episodes spanning 14 datasets, 11 encoders, and three modalities, TabPFN achieves the best mean rank among nine classification heads on both negative log-likelihood (NLL) and expected calibration error (ECE). At a representative setting, it reduces NLL by 48--62% and ECE by 2.1--5.3$\times$ relative to the average of the eight baselines while matching or exceeding their average accuracy. Its accuracy advantage is conditional, concentrating at moderate-to-high shot counts and low-to-moderate feature dimensions ($k \ge 50$, $d \le 32$), and diminishing when labeled data are scarce, feature dimensions are high, or competing methods approach ceiling accuracy. In targeted backbone-adaptation experiments, replacing the trained linear head with TabPFN substantially improves calibration while preserving competitive accuracy. These results provide empirical guidance for using TabPFN as a training-free head in calibration-sensitive multimodal classification. To support transparency and reproducibility, we publicly release the source code, experiment configurations, and evaluation scripts in our GitHub repository: https://github.com/Jingxiang-Zhang/tabpfn-multimodal-embeddings.

Fully characterizing a crystalline material requires integrating heterogeneous data sources -- atomic structures, diffraction patterns, electronic density of states, and natural language -- each of which captures a different facet of the same physical object. In practice, however, these modalities are stored and analyzed in isolation, making it difficult to relate or query materials across representational boundaries. We present MatBind, a contrastive learning framework that aligns four materials modalities -- crystal structure, powder X-ray diffraction (pXRD) simulated from structures, density of states (DOS), and text -- into a unified embedding space using crystal structure as the central physical anchor. The framework induces alignment between modalities never explicitly paired during training, enabling emergent zero-shot cross-modal retrieval as a direct consequence of the shared representation. The learned embedding space organizes materials according to physically meaningful properties without explicit supervision, and retrieval performance improves systematically when modalities are combined at query time. These results demonstrate that treating heterogeneous materials data as complementary projections of a single physical reality, rather than as isolated data sources, is not a practical choice but is consistent with the underlying physics.

Multimodal learning usually requires a dedicated encoder per modality. When a tabular modality is involved, prior work has been mostly using a plain MLP as the encoder. Yet if it were a strong encoder, the tabular domain would not be ``the last unconquered castle for deep learning''. This study evaluates state-of-the-art tabular models as encoders in the image-tabular setting for the first time. An obstacle stands out. In-Context Learning models, among the best performing methods in the tabular domain, require labels to process instances, making it non-trivial to embed training and test instances the same way. We addressed this problem across multiple models of this family. With this study, we would like to highlight the importance of encoder factor in the multimodal learning.

While generative models enable encoding of complex neuroimaging data for feature generation and reconstruction, developing optimal architectural frameworks with appropriate encoding and latent space processes is crucial for studying structural and functional properties of the brain. We design a multimodal generative framework for structural and functional magnetic resonance imaging (MRI) features through systematic evaluation of encoding strategies, latent multimodal fusion, and generative model selection. Using structural gray matter volume (GMV) and static functional network connectivity (sFNC) features from a large neuroimaging dataset, we analyze generative frameworks involving variational autoencoders (VAEs), transformers, generative adversarial networks (GANs), and diffusion models. Architectures that employ modality-aware graph encoding of functional connectivity into a lower-dimensional latent space outperform vectorized encoders or direct data space approaches. The proposed multimodal graph VAE (gMMVAE) surpasses alternative generative variants across multiple metrics for generation fidelity, reconstruction quality, efficiency, and latent space discriminability, highlighting its potential for robust multimodal neuroimaging analysis.

We present VTMR, a two-stage framework for Video-To-Music Recommendation. In Stage1, VTMR aligns comprehensive video and music signals in a joint audio-visual-text representation space and efficiently retrieves semantically compatible candidates using coarse global embeddings. In Stage2, it reranks the retrieved candidates by attending to the temporal sequences of both video and music, thereby capturing fine-grained temporal correspondence. Evaluated on the video-to-music recommendation task, the multimodal retrieval stage improves R@10 from 14.2 to 15.9 and Median Rank from 75 to 58 over the strongest baseline; the temporal reranker further boosts R@10 to 18.3 and Median Rank to 46, demonstrating complementary gains from richer query encoding and temporal alignment. A human preference study confirms that VTMR is on par with a commercial baseline in overall preference, while outperforming a generative baseline in music quality.

Multimodal document retrieval aims to retrieve relevant pages while preserving both textual and visual content from the original document. However, existing benchmarks primarily evaluate simple lexical or semantic matching, and most methods encode pages independently. Consequently, they overlook the contextual information in the document required to resolve queries that aggregate information across multiple pages. In this paper, we introduce CMDR and CMDR-Bench, a new multimodal document retrieval task and benchmark that require modeling document context. To address this challenge, we propose CMDR-Embed, a contextual multimodal embedding framework that explicitly incorporates document context by jointly encoding multiple pages and deriving page-level embeddings from a shared contextual representation. Furthermore, we introduce CMCL, a contextual multimodal contrastive learning objective that effectively trains CMDR-Embed by balancing contextual modeling with page-level discriminability. Experiments demonstrate that CMDR-Embed significantly outperforms non-contextual embeddings, highlighting the importance of context-aware multimodal embeddings for advancing document retrieval.

While garments are essential for realistic digital humans, their topological variety makes them much harder to model than parametric bodies. Traditional tailoring relies on 2D sewing patterns, yet bridging these patterns to 3D geometry currently requires physical simulations. We present Stitched Embeddings, the first simulation-free framework to unify 3D garment reconstruction and sewing pattern inference within a single bidirectional latent space. By leveraging the geometric priors of a pretrained 3D foundation model, our approach overcomes the data scarcity typically associated with high-quality garment modeling. We propose to use the BoxMesh as a critical intermediate representation to align 2D panels into 3D configurations without the computational overhead of a simulator. This architecture achieves state-of-the-art accuracy in pattern reconstruction while significantly improving efficiency. Furthermore, our differentiable pipeline enables novel applications, including pattern recovery from meshes and 3D editing from 2D patterns. Finally, this work provides a scalable link between neural 3D vision and the physical garment manufacturing pipeline. Project Page: https://andreus00.github.io/stitchedembeddings

This paper offers an in-depth analysis of non-sequential multimodal sentence-level embeddings, with a particular focus on the SONAR model. We demonstrate that certain embedding dimensions are sensitive to perturbations and can serve as indicators of decoding anomalies. By leveraging the consistency between successive encoding and decoding, we successfully build an accurate detector. Additionally, we explore modifying specific dimensions of interest to attempt to correct them. This work underscores the importance of understanding and analyzing the embeddings themselves to enhance the reliability of multimodal representations.

Affective computing increasingly relies on deep learning to represent emotions, yet latent spaces often remain opaque, high-dimensional black boxes. This paper investigates whether Transformers' embeddings recover the geometric regularities of Russell's circumplex model. We unify two complementary experiments testing the hypothesis that, after training models on text and speech, their resulting latent spaces encode a topology consistent with valence-arousal and reproduce human-like neighborhood relations. Specifically, we evaluate deep representations extracted from Transformer-based text (RoBERTa) and speech (wav2vec 2.0) encoders, along with a multimodal Transformer fusion architecture, across naturalistic datasets like MSP-Podcast and controlled LLM-generated stimuli. Our analysis reveals that multimodal fusion of text and audio yields perfect topological alignment with Russell's primary emotion ordering. Furthermore, in a zero-shot setting using generic text embeddings, projected fine-grained emotion terms fall close to their established human-mapped coordinates. Our contribution is a novel, data-driven framework for validating emotion models, demonstrating that Russell's circumplex structure is intrinsically encoded in the embeddings of these modalities rather than being solely an artifact of human labeling, thereby bridging the gap between psychological theory and representation learning.

Semantic search and recommendation of similar documents, such as news and reports about unusual environmental events (e.g., a dead whale washed ashore in Alaska) that contain spatial and temporal information, is a critical task in Geographic Information Retrieval (GIR). This work presents a novel framework that leverages AI foundation models, including Large Language Models (LLMs) and Vision-Language Models (VLMs), to enable effective similarity search and ranking for such event documents. To support this goal, we introduce two new strategies: (1) CAMERA (Context-Aware Multimodal Event Retrieval Algorithm), which fuses textual and visual information to generate richer embeddings than those derived from text alone; and (2) ASTRA (Adaptive Spatial and Temporal Re-ranking Algorithm), which improves similarity ranking by incorporating scale-dependent spatiotemporal relevance alongside semantic similarity. Experimental results, using a dataset from the Local Environmental Observer Network, demonstrate that our VLM-enhanced methods outperform unimodal, LLM-based approaches in similarity ranking effectiveness. By automatically linking relevant event reports, the proposed framework helps both data curators and the general public gain deeper insights into environmental change and its localized impacts. These findings highlight the potential of AI foundation models to advance GIR through multifaceted, intelligent analysis that integrates key geographic concepts: space, time, scale, and semantics.

We introduce the Massive Video Embedding Benchmark (MVEB), a 23-task benchmark for video embeddings spanning classification, zero-shot classification, clustering, pair classification, retrieval, and video-centric question answering. We evaluate 33 models and find that no single model dominates: MLLM-based embeddings lead on classification, clustering, pair classification, and QA; multimodal binding leads on retrieval and zero-shot classification; generative MLLMs without contrastive adaptation collapse on cross-modal tasks. Paired video-only vs. audio+video evaluations show that audio's contribution depends on dataset annotation provenance: audio helps when labels were produced from both modalities and hurts when they were produced from visuals alone, a six-point gap consistent across model families. MVEB is derived from MVEB+, a 184-task pool, and is designed to maintain task diversity while reducing evaluation cost. It integrates into the MTEB ecosystem for unified evaluation across text, image, audio, and video. We release MVEB and all 184 tasks along with code and a leaderboard at https://github.com/embeddings-benchmark/mteb.

Reasoning-driven universal multimodal embedding has advanced rapidly by introducing Chain-of-Thought (CoT) reasoning into the embedding pipeline. Despite the strong performance across both general and complex tasks, this paradigm suffers from two core limitations: (i) autoregressive CoT reasoning incurs high computational cost, making it impractical for low-latency retrieval; and (ii) embedding performance is heavily coupled with CoT annotation quality, making large-scale training unreliable. These raise fundamental questions: Is textual CoT the optimal form of reasoning for embedding, and can effective embedding reasoning be accomplished in latent space? To this end, we propose LaME (Latent Reasoning Multimodal Embedding), which formulates embedding-oriented latent reasoning as a weakly supervised information bottleneck. LaME employs K learnable reason tokens as a fixed-capacity bottleneck, completing all reasoning within a single forward pass. The two weak supervision signals structurally decouple contrastive from autoregressive objectives and eliminate dependence on CoT annotations, while a two-stage training pipeline ensures stable convergence. Experiments on MMEB-v2 and MRMR show that LaME achieves competitive performance, surpassing some explicit CoT-based models, while delivering 60x faster inference than explicit CoT methods and 2x faster than latent baselines with throughput comparable to discriminative embedding models. Code is available at https://github.com/PeppaWu/LaME.

Clustering is fundamental to scRNA-seq analysis, serving as a cornerstone for identifying cell populations and resolving tissue heterogeneity. However, existing methods focus on mining numerical statistical patterns, suffering from semantic agnosticism by neglecting the intrinsic biological functions encoded by genes. While Large Language Models (LLMs) offer promising semantic capabilities, their direct adaptation to cell clustering is hindered by the structural mismatch between generative pre-training objectives and discriminative downstream tasks. To bridge this gap, we propose scLLM-DSC, a novel LLM-Knowledge Enhanced Cross-Modal Deep Structural Clustering framework. Diverging from data-driven paradigms, scLLM-DSC establishes a semantically-grounded representation by synergizing two views: a Knowledge-Driven Semantic View derived from NCBI gene priors and contextualized Cell2Sentence embeddings, and a Structure-Aware Topological View extracted via a graph-guided encoder. Crucially, we introduce a cross-modal contrastive alignment mechanism to enforce consistency between biological semantics and transcriptomic features within a unified latent space. Extensive benchmarks demonstrate that scLLM-DSC significantly outperforms eleven state-of-the-art baselines in clustering accuracy.

Audio-visual Generalized Zero-shot Learning (AV-GZSL) is a challenging task that aims to classify both seen and unseen objects or scenes by integrating data from audio and visual modalities. Recent studies primarily focus on fusing or aligning audio and visual features to generate more informative audio-visual embeddings. Also, aligning the audio-visual and textual features of most existing methods relies solely on the optimization objectives. However, those methods neglect the inherent distributional and structural differences between audio-visual and textual modalities. To address this limitation, we propose a method termed Aligning Hierarchical Standardized Embedding (AHSE), which enables hierarchical alignment of standardized audio-visual and textual embeddings within a shared embedding space. Specifically, we first apply Z-score standardization to the fused audio-visual and textual embeddings to reduce distributional mismatches. We then introduce a hierarchical alignment strategy that minimizes discrepancies at the semantic, class, and batch levels, thereby constructing a more robust and well-structured embedding space. This strategy not only preserves semantic and inter-class relationships but also maintains spatial consistency within each batch. Extensive experiments on three benchmark datasets: VGGSound-GZSL, UCF-GZSL, and ActivityNet-GZSL, demonstrate that AHSE achieves competitive performance in zero-shot learning.

Omni-modal retrieval promises a single embedding space for text, image, video, document, and audio inputs, but building such a unified retriever is difficult since these modalities differ in data distribution, architecture, and optimization dynamics. In this work, we present Conan-embedding-v3, a decouple--fuse--recover framework for omni-modal retrieval. Conan-embedding-v3 first trains modality specialists independently and fuses their task vectors into a single dense backbone, a strategy we call Decoupled Specialist Fusion. We show that this fusion composes visual, video, and document retrieval capabilities, but also exposes a failure mode for projector-based modalities: when audio is attached through an external encoder and projector, fusing the backbone leaves the projector calibrated to the audio-specialist backbone, causing a large audio retrieval regression despite copying all audio-specific modules unchanged. We call this failure Projector Drift. To repair it, Conan-embedding-v3 applies Projector Recovery (i.e., full-parameter fine-tuning of the projector while keeping the backbone frozen) followed by balanced multi-modal rehearsal. The resulting model supports these retrieval pathways in one backbone, achieving 74.9 scores on MMEB while obtaining 55.61 on the 30-task MAEB audio suite.

Recent advancements have significantly expanded the theoretical context windows of Multimodal Embedding Models (MEMs). However, larger context windows do not necessarily translate into effective comprehension and representation of long-context multimodal inputs, which remains a critical bottleneck for real-world deployment. To address the lack of systematic evaluation in this setting, we introduce MMLongEmbed, the first comprehensive benchmark for evaluating MEMs in long-context scenarios. MMLongEmbed comprises four retrieval tasks spanning multiple context-length ranges, covering text, document, and video modalities. Through extensive evaluation of state-of-the-art models, we find that current architectures rely heavily on superficial feature matching and struggle to capture deep semantic and structural dependencies. We further observe that performance degradation varies systematically with context length and key information placement. Moreover, models exhibit substantially different robustness to redundant contextual information across modalities. For reproducibility, the benchmark and code are publicly available.

Retrieval over visually-rich documents, pages that interleave text with figures, tables, and charts, is essential for multimodal retrieval-augmented generation, yet most retrievers still discard the visual channel. The Multimodal Document Retrieval Challenge, Track~1 of the MIR Challenge at the first EReL@MIR workshop, co-located with The Web Conference 2025, asks participants to build a single retrieval system that handles two complementary regimes: closed-set document page retrieval within long documents from a text query (MMDocIR), and open-domain retrieval of Wikipedia-style passages from an image or image-plus-text query (M2KR). Systems are ranked by the macro-average of mean Recall@$\{1,3,5\}$ over the two tasks. The challenge drew 455 entrants and 586 submissions across 22 teams. This report describes the challenge design, datasets, and evaluation protocol; reports the final standings; and analyses the three winning teams' systems. All three build on decoder-based Multimodal-LLM embedders from the Qwen2-VL family rather than on CLIP-style encoders, and differ chiefly in whether they reach the top through fine-tuned ensembles, training-free multi-route fusion with a strong vision-language re-ranker, or zero-shot late interaction. The training-free system finished within $0.1$ point of the fine-tuned winner.

This work presents MLLM-Microscope, a novel system designed for analyzing the hidden representations within Multimodal Large Language Models (MLLMs). Our system evaluates the linearity, intrinsic dimension, and anisotropy of multimodal token embeddings across transformer layers. Utilizing the ScienceQA dataset, we evaluate two state-of-the-art MLLMs, LLaVA-NeXT and OmniFusion. We find that both the main and residual streams for tokens of both modalities exhibit highly linear behaviors across transformer layers. However, LLaVA-NeXT's image tokens reveal a slight decline in linearity, whereas OmniFusion's remain consistent. Image token dimensions in OmniFusion remain consistently higher across layers compared to LLaVA-NeXT. Also, the OmniFusion's anisotropy is observed to stay consistently low throughout the layers. These findings suggest that the inner workings of MLLMs highly depend on the nature of modality fusion performed before passing the token sequence into LLM. This and other new potential insights obtainable from our system are surely capable of enhancing our understanding of the inner workings of MLLMs, informing future model design and optimization.

Protein-protein interactions (PPIs) are essential for many biological processes. However, existing PPI prediction approaches suffer from two major limitations: they overlook the hierarchical organization of proteins, particularly meso-scale motifs that critically regulate PPIs, and fail to effectively integrate sequence, structure, and function modalities. To address these limitations, we propose MMM-PPI, a Hierarchical Motif-based Multi-Modal protein Encoder for PPI Prediction that constructs PPI embeddings in a bottom-up multi-modal manner across three scales. At the micro-scale, we encode three modal residue features; at the meso-scale, a novel multimodal motif encoder aggregates residues into spatially-informed motif embeddings; at the macro-scale, a multimodal protein encoder integrates motifs into protein embeddings by jointly modeling motif importance and inter-modal correlations. The pre-trained encoder can be used off-the-shelf for large-scale PPI prediction. Extensive experiments on multiple PPI datasets show that MMM-PPI outperforms state-of-the-art multi-label PPI prediction models, particularly under challenging data partitions and limited data scenarios. Codes are in https://github.com/yzf-code/MMM-PPI.

Transformer-based architectures have advanced sequence modeling in language and vision, yet general-purpose representation learning for heterogeneous multivariate time series remains underexplored. We introduce CHARM (Channel-Aware Representation Model), which incorporates channel-level textual descriptions into a Transformer encoder equivariant to channel order. CHARM is trained with a Joint Embedding Predictive Architecture (JEPA) and a novel loss promoting informative, temporally stable embeddings; latent-space prediction encourages robustness to sensor noise while description-aware gating provides interpretability through learned inter-channel relationships. Across anomaly detection, classification, and short- and long-term forecasting, the learned embeddings achieve strong performance using only a linear probe. Performance is driven primarily by the JEPA objective and conditioning architecture, with text descriptions serving as channel identifiers for cross-dataset generalization.

Multimodal models in oncology can produce accurate predictions, but accurate prediction does not reveal whether the model has learned biology that is shared across modalities, biology confined to one modality, or spurious correlations that reflect confounders rather than genuine biology. We introduce DECAT, a model-agnostic post-hoc evaluation framework that classifies multimodal representations into four diagnostic scenarios for a given task and modality, using five null-referenced metrics and a rule-based decision procedure. The framework operates on learned representations, requires no knowledge of which specific confounder is present, and returns indeterminate when the evidence is insufficient. We validate DECAT on synthetic data across four multimodal model classes (over 2,500 trained representations) and on real data from 8,979 TCGA patients, evaluating both multimodal embeddings and five pretrained pathology foundation models. Entangled models (e.g., CLIP) achieve near-perfect shared biology detection but falsely claim shared biology in the majority of cases where it is absent on real foundation model embeddings. This false claim rate increases with confound strength so that larger cohorts and stronger representations produce more confident but still incorrect diagnoses. Applied to both multimodal TCGA embeddings and five pathology foundation models without paired RNA, DECAT detects confounding invisible to AUROC without requiring the confounder labels, as confirmed by post-hoc stratification.