Multimodal Embedding

Follow this topic Ask about this topic

Multimodal embedding studies how to represent different kinds of data so that related items are close in a shared space, enabling tasks such as cross-modal search, recommendation, and classification. Common approaches include contrastive dual encoders, instruction-tuned multimodal language models, and alignment methods based on graphs or optimal transport; multi-vector and late-interaction representations can retain finer detail than a single vector. Benchmarks such as MMEB, ViDoRe, and audio- and video-retrieval suites test these models across tasks and modalities, while newer methods explore compressed representations and latent reasoning to improve quality without costly inference. A central challenge is balancing broad, zero-shot capability with fine-grained, task-specific information, while keeping representations efficient and robust when modalities are missing or unevenly informative.

Latest papers

160 papers19 papers in the last four weeks, up 6% on the four weeks before. 0.1% of all new papers.

Oct 27, 2026

EmbeddingGemma 2: Unified and Efficient Multimodal Representations

We present EmbeddingGemma 2, an open multimodal embedding model which natively represents text, images, audio, video, and combinations thereof in a unified space. Our approach introduces improvements such as simplified per-layer embeddings and a multi-stage curriculum with modality and objective progression which ensure strong quality across unimodal, cross-modal, and mixed-modal evaluations. Notably, in comprehensive embedding benchmarks across all modalities, EmbeddingGemma 2 establishes a new Pareto frontier, outperforming both specialist models and models two to twelve times larger. Furthermore, our design offers flexibility for resource-constrained environments, supporting selective modality encoder loading, adjustable vision token budgets, truncatable embedding dimensions, and quantized weights. We present ablations analyzing key decisions and release EmbeddingGemma 2 to promote further research.

Henrique Schechter Vera, Sahil Dua, Jovan Andonov +70

Multimodal Embedding Unified Multimodal Models Multimodal Pretraining Efficient Multimodal Inference

Oct 23, 2026

Beyond Post-Hoc Rewards: Privileged Reasoning Exploration for Multimodal Embeddings

Advances in Multimodal Large Language Models (MLLMs) have driven substantial progress in multimodal embeddings. Recent methods exploit generative capabilities by augmenting embedded inputs with self-generated reasoning to produce more informative embeddings. However, conventional retrieval supervision for training embedding models specifies pairwise relevance relations without explicitly guiding the reasoning used for input augmentation. Such reasoning is neither directly observed nor uniquely determined by the input, allowing multiple plausible traces to introduce semantic cues of varying retrieval utility. Existing approaches largely optimize reasoning conditioned on the input alone, limiting the exploration of retrieval-relevant traces. We introduce PILOT, a framework that models reasoning as a latent variable and instantiates its prior and posterior distributions as distinct reasoning modes of a shared MLLM. The prior conditions only on the input, whereas the posterior uses additional training-time relational evidence to guide reasoning exploration. To translate the posterior's broader exploration into inference-time improvements, we derive an importance-weighted variational objective and instantiate it as a prior-posterior co-evolution framework that improves both modes under a shared inference-compatible reward while selectively transferring posterior trajectories to the prior according to their utility and prior reachability. On the 78-task MMEB-V2 benchmark, PILOT outperforms the primary baseline across all 12 task groups at both 2B and 7B parameters, improving overall scores by 3.2 and 3.3 points, respectively, with the largest gains in visual-document retrieval (+6.1 and +6.4 points). Our results show that relational supervision is most valuable not as a post-hoc reward but as a signal for shaping how reasoning is explored in training reasoning-enhanced embedding models.

Tianyu Yang, Yijun Tian, Lars Benedikt Kaesberg +4

University of Göttingen · University of Notre Dame · State and University Library of Göttingen

Multimodal Embedding Cross-Modal Retrieval Multimodal Reasoning

Oct 23, 2026

OlmoEarth v1.3: Embeddings

We make some OlmoEarth embeddings. To do this, we train a new model: OlmoEarth v1.3. OlmoEarth v1.3 is designed to condense multimodal, multitemporal remote sensing data into a single digestible 64 or 128-dimensional vector. OlmoEarth v1.3 produces state-of-the-art embeddings. In addition, OlmoEarth v1.3 is a strong encoder backbone -- competitive with OlmoEarth v1 and v1.2 -- while requiring 18% and 53% of the GPU hours to train, respectively. We use OlmoEarth v1.3 to compute global, annualized 10m embeddings, which we release under the OlmoEarth artifact license.

Gabriel Tseng, Favyen Bastani, Patrick Alan Johnson +13

Allen Institute for AI

Geospatial Representation Learning Multimodal Embedding Remote Sensing Temporal Representation Learning Multimodal Pretraining

Oct 21, 2026

pplx-embed-v2-late Technical Report

In this report, we introduce pplx-embed-v2-late, a pair of 0.6B and 9B multilingual and multimodal late-interaction models. Both models achieve state-of-the-art results within their respective size classes across a range of text, image, and visual document retrieval benchmarks. On visual document retrieval tasks, pplx-embed-v2-late-0.6b is competitive with models several times larger than itself. The models are trained via representation distillation from an 18B teacher. Consequently, they share an embedding space that supports asymmetric deployment: documents can be indexed with the 9B model and queried with the low-latency 0.6B model. Both models are available at https://huggingface.co/collections/perplexity-ai/pplx-embed-v2.

Markus Krimmel, Louis Milliken, Ekaterina Glazkova +8

Perplexity AI

Multimodal Embedding Multimodal IR Late Interaction Retrieval Visual Document Retrieval

Oct 16, 2026

Inducing Fourier Geometry: A Smoothness Prior Imposes Approximate Translation Equivariance in Multimodal Stellar Embeddings

Learned representations often contain geometric structure induced by symmetries in the underlying data. We ask whether a known continuous parameter space can instead be used to induce such structure when the corresponding symmetry is only approximate. We introduce a graph-Laplacian smoothness prior that encourages representations to vary smoothly over a physical parameter space without specifying a representation geometry; in the continuum limit it selects Laplacian eigenfunctions and therefore predicts the Fourier-like geometry associated with translation symmetry. On multimodal stellar representations spanning six astronomical surveys, the prior induces Toeplitz Gram structure and sinusoidal modes along stellar parameters, including in modalities whose native representations lack this geometry, while modality-specific residuals preserve structure outside the imposed coordinates. Of the exact-symmetry predictions, stationarity in a recoverable warped coordinate survives; diagnostics assuming exact quantization or asymptotic rates do not. The prior also improves downstream linear-probe performance and label efficiency, suggesting a general strategy: where a meaningful continuous parameter space exists, its local geometry can serve as a data-informed prior.

Ilay Kamai, Alex M. Bronstein, Hagai B. Perets

Technion - Israel Institute of Technology · ISTA - Institute of Science and Technology Austria

Geometric Representation Learning Equivariant Representation Learning Multimodal Embedding

Oct 16, 2026

Domain-specific language alignment of multimodal embeddings towards explainable predictions

Multimodal data are commonly fused into latent representations that improve downstream task performance but are not interpretable, limiting their application in domains that require explainable model decisions. Vision-language alignment has made image embeddings interpretable through language, but requires vast image-caption training data. However, captioned data are typically not available for multimodal embeddings, as relevant features cannot directly be observed and annotated from embeddings alone. Here, we propose a framework to language-align any embedding space that can be linked to interpretable auxiliary data. Captioning functions convert auxiliary data into partial captions; a soft contrastive loss weights pairs by the similarity of their properties that each caption describes; and a rank-based concept retrieval task evaluates alignment against the auxiliary data rather than the captions. We apply this framework to multimodal Earth embeddings to facilitate explainable predictions for two species distribution modelling tasks: S2BMS and SatBird. AlphaEarth embeddings beat previous state-of-the-art models on 6/7 metrics, while our framework language-aligns domain-specific concepts of the embedding space, outperforming previous Earth observation vision-language models on average concept retrieval for both datasets. Finally, aligned concepts recover species-habitat associations (

ρ

=0.44) and relate habitat descriptions to species predictions, showing that abstract embeddings can be interrogated in the language of domain experts.

Gabrielė Tijūnaitytė, Thijs L. van der Plas, Mohamed-Bachir Belaid +3

Artificial Intelligence Group, Wageningen University & Research, Wageningen, the Netherlands · Machine Learning in Earth Observation Lab, University of Bonn, Germany · NILU, Climate and environmental research institute, Kjeller, Norway +1

Multimodal Embedding Concept-Based Explanations Vision-Language Alignment Remote Sensing

Oct 16, 2026

VTrajEmbed: Aligning Text with Video Object Trajectories via Spatiotemporal Embeddings

Fine-grained alignment between text and video object trajectories is a core challenge in multimodal understanding. Existing embedding models excel at global retrieval, yet their single global representation entangles multiple objects and their spatiotemporal dynamics, limiting direct object-level grounding. To address this issue, we propose VTrajEmbed, an embedding model that decomposes a video into object trajectories and aligns each trajectory embedding with text for object-level grounding through retrieval. Our main contributions: (1)Slow-Fast Trajectory Encoding, a dual-branch architecture that establishes trajectory representations by densely sampling low-resolution frames for high-frequency motion cues and sparsely selecting high-resolution frames for scene context, aggregated by a shared LLM backbone; (2)Mask-Guided Contrastive Learning, a training objective built upon these representations that couples an intra-group loss and a batch-level bidirectional loss under a relation mask, excluding unverified relations and preventing confirmed matches from becoming false negatives; (3)Text2Traj-Bench, a trajectory-level text-video benchmark with model-assisted construction and human-verified query--trajectory relations. Experimental results show that VTrajEmbed-5B improves text-to-trajectory and trajectory-to-text R@1 over Qwen3-VL-Embedding-8B by 15.48 and 18.05 percentage points, respectively, and achieves 49.48% average R@1 across five global retrieval benchmarks, compared with 42.86% for Qwen3-VL-Embedding-8B.

Ruizhi Chen, Zining Fan, Yudi Guo +5

Xingchen AGI Lab, China Telecom Artificial Intelligence Technology (Beijing) Co., Ltd. · Peking University · Nanjing University +1

Multimodal Embedding Vision-Language Retrieval Contrastive Learning

Oct 15, 2026

TWIN-MoE: Two-stage Intra-Expert Imputation Network for Incomplete Multimodal Learning

Recent Mixture-of-Experts (MoE) approaches address missing modalities either by routing each missingness pattern to a dedicated expert, which scales poorly as the number of modalities grows, or by imputing the missing modality from representations contributed by multiple experts, which requires a costly retrieval mechanism. We show that routing induces expert-conditioned representation neighborhoods and derive an imputation-error bound controlled by their local feature dispersion. When these expert-conditioned neighborhoods are more compact than the global representation space, intra-expert aggregation admits a tighter worst-case reconstruction bound than unconstrained global aggregation. Guided by this analysis, we propose TWIN, a two-stage intra-expert imputation framework. Pre-imputation generates a context-conditioned estimate in the encoder embedding space and radially soft-clips it using modality-specific statistics maintained in an observed FIFO queue, this ensures that the pre-imputed modality stays in empirically calibrated region of the target modality space. Post-imputation refines the missing representation after MoE by aggregating expert-conditioned observed representations within the same sample. Across four missing-modality benchmarks under natural, random, and asymmetric dropout protocols, TWIN achieves state-of-the-art performance while using substantially fewer parameters than competing MoE-based imputation methods.

Liang Nathan Zheng, Jiyuan Liu, Wenhao Liang +3

School of Computer Science and Information Technology, Adelaide University, Adelaide, SA 5000, Australia

Missing-Modality Learning Multimodal Learning Mixture-of-Experts Models Multimodal Embedding

Oct 15, 2026

Stuck on Shuffle: On Attribute-Object Binding in Vision-Language Models

Retrieval is a fundamental task in multimodal systems: given an image or text query, the goal is to find the most relevant images or texts in a collection, often using a shared embedding space. Reliable retrieval requires these embeddings to capture not only which concepts are present, but also how they are composed. One key compositional challenge is attribute--object binding: encoding which attributes belong to which objects. Existing benchmarks provide a limited view of this capability, typically covering a subset of modality pairings, simple two-object swaps, or small attribute sets. Consequently, it remains unclear how robustly current models preserve attribute--object bindings across different modality pairings and more complex settings. We therefore introduce SABER, a controlled benchmark for evaluating attribute--object binding across all four image and text modality pairings, spanning diverse attribute types and scenes with up to ten objects. Evaluating models on SABER reveals substantial binding failures in both CLIP-style dual encoders and recent MLLM-based embedders. These findings motivate us to ask how binding can be improved efficiently. While multimodal supervision may be effective, collecting suitable images is difficult and costly. Text, on the other hand, is well suited to binding supervision: it is cheap to generate, supports a broader range of attributes and enables precisely controlled hard negatives. We thus finetune an MLLM-based embedder exclusively on synthetic attribute-binding text. Despite the absence of image supervision, the gains transfer to cross-modal and image--image retrieval while preserving general multimodal embedding performance. Our model achieves SOTA results on SABER while also improving on negation and instance-retrieval benchmarks, demonstrating broader benefits of our binding-focused supervision.

Netanel Y. Tamir, Michael Green, Issar Tzachor +4

OriginAI · NVIDIA · Tel Aviv University

Vision-Language Retrieval Multimodal Embedding Multimodal Model Evaluation

Oct 15, 2026

CIM-Embed: Learning Instruction-Conditioned Multimodal Embeddings for Clustering

Multimodal objects can require different partitions under different instructions, as when products in an e-commerce catalog are grouped by color, material, or both. Retrieval-oriented embeddings may not recover these partitions, while generative clustering often relies on intermediate descriptions or iterative decisions. This paper presents CIM-Embed, a framework for learning instruction-conditioned multimodal embeddings for clustering. Supervised contrastive learning uses same-group positives and negatives from other groups, optionally including other tasks; heterogeneous structured batches accommodate tasks with different group counts. At inference, the model maps each object's image, text, and instruction directly to an embedding; HDBSCAN then forms a partition without the ground-truth cluster count. We construct ProductMIC from product metadata in Amazon Reviews 2023, comprising training data and an evaluation benchmark for instruction-conditioned clustering. The benchmark contains 5,373 products and 449 single-attribute and compositional grouping tasks spanning 50 leaf categories excluded from our training. With full-dimensional embeddings and HDBSCAN, direct within-task SupCon training achieves a mean per-task V-measure of 95.01, compared with 53.06 for WeMM 9B, the strongest general-purpose embedding baseline evaluated. Frozen product-trained encoders also improve out-of-domain conditional grouping, while controlled transfer experiments distinguish partition transfer from instruction response. We will release the code and dataset upon acceptance.

Yinghua Xu, Pengyang Xie, Yanxiang Zhou +4

College of Control Science and Engineering, Zhejiang University · Alibaba International Digital Commerce

Clustering Multimodal Embedding Multimodal Contrastive Learning

Oct 10, 2026

What Really Works for Universal Multimodal Embeddings? A Controlled Study of Contrastive Training Strategies

Contrastive learning has become the standard paradigm for multimodal representation learning, aligning semantically matched pairs while separating mismatched ones in a shared embedding space. Although various contrastive training strategies have been proposed across negative selection, batch construction, and gradient modulation, their actual effectiveness under controlled, reproducible conditions remains poorly understood. This motivates a fundamental question: Do sophisticated contrastive strategies consistently bring meaningful downstream gains in multimodal embedding learning? To address this, we systematically evaluate representative strategies under a unified benchmarking framework on MMEB. Surprisingly, direct evaluations reveal that most existing strategies offer marginal gains over a standard aligned baseline, with several even degrading overall performance. To uncover the underlying failure modes, we introduce an MLLM-as-a-Judge diagnostic framework that dissects top-1 retrieval behaviors into labeled positives, false negatives (unlabeled valid matches), and hard/easy negative errors. Our systematic analysis yields three key insights: (1) Gradient distortion: Upweighting gradients from high-similarity negatives consistently hurts downstream generalization; (2) Distribution illusion: Wider separation between positive and negative similarity distributions does not necessarily translate to better retrieval performance; and (3) Synergistic training: While false-negative mitigation or hard-batch construction yields limited gains in isolation, integrating hard-negative mining, false-negative mitigation, and hard-batch construction achieves optimal performance, delivering an average improvement of 6.0% on MMEB with Qwen2-VL-2B. Overall, our findings demystify popular contrastive strategies and provide actionable recipes for training universal multimodal embedding models.

Duo Li, Haojun Pan, Zuhao Yang +4

CCDS, Nanyang Technological University · Alibaba Group · Universität Hamburg +1

Multimodal Embedding Multimodal Contrastive Learning Cross-Modal Retrieval Information Retrieval Multimodal Model Evaluation

Oct 10, 2026

EarthAlign: Learning Language-Aligned Representations of Earth System Data

Interpreting Earth system conditions is central to understanding environmental risks and informing decisions across weather- and climate-sensitive sectors. Existing multimodal weather systems primarily target task-specific prediction, generation, or question answering, while general-purpose alignment between Earth system data and language remains underexplored due to the limited availability of diverse language supervision grounded in meteorological conditions. To fill this gap, we introduce EarthAlign, a large-scale dataset with broad geographic coverage spanning 2000-2024 that combines 10 expert-curated natural event and disaster data sources with climatology-derived normal-weather samples. Each sample includes temporal and geospatial information and four grounded caption styles aligned with 63-variable ERA5 meteorological data. Building on EarthAlign, we introduce MeteoLIP, a dual-encoder framework for learning aligned meteorology-language representations. MeteoLIP incorporates MetTok, which uses a geolocation- and time-conditioned query to aggregate variable-specific patch embeddings into a compact sequence of meteorology tokens. To comprehensively assess Earth system multimodal representations, we introduce three downstream tasks: cross-modal retrieval, weather-condition classification, and physical variable estimation. Our experiments show that MeteoLIP improves upon the corresponding CLIP and SigLIP baselines across these tasks. Further analyses highlight the importance of meteorological input, temporal and geospatial conditioning, and multi-style supervision for meteorology-language alignment. Thus, EarthAlign and MeteoLIP provide a foundation for learning and evaluating language-aligned representations of Earth system data. Our code, dataset, and weights are publicly available at https://github.com/mbzuai-oryx/EarthAlign.

Fadillah Adamsyah Maani, Numan Saeed, Jiaqi Ma +4

Mohamed bin Zayed University of Artificial Intelligence · ADIA Lab · University of California San Diego

Cross-Modal Representation Learning Multimodal Embedding Earth Science Cross-Modal Retrieval

Oct 8, 2026

Syn-Omni: Structured Specialization and Progressive Collaboration for Omnimodal Embeddings

Omnimodal embeddings naturally involve both shared representations and modality-specific features across heterogeneous inputs. However, existing omnimodal embedding methods often rely on a single shared parameter space over mixed-modality data, limiting structural separation between universal and modality-specific representations. To address this, we propose Syn-Omni, a unified framework for structured omnimodal adaptation with modality specialization and controlled cross-modal collaboration. Specifically, we introduce Orthogonal Modality-Expert LoRA (OME-LoRA), which decomposes adaptation into a shared LoRA path for universal semantics and modality-expert LoRA paths for modality-aware specialization. Furthermore, Progressive Synergy Routing (PSR) enables experts to first establish modality-specific priors, then gradually interact with other modality-experts for cross-modal synergy. Evaluated across 81 diverse tasks spanning image, video, audio, and audiovisual modalities, Syn-Omni consistently outperforms omnimodal baselines, demonstrating the effectiveness of structured specialization and cross-modal progressive collaboration.

Youngtaek Oh, Qiyu Wu, Hiromi Wakaki +2

KAIST · Sony Group Corporation

Multimodal Embedding Cross-Modal Representation Learning Mixture of Experts Low-Rank Adaptation

Oct 8, 2026

Autoregressive Retriever: Improving Query Understanding from Item Feedback for Universal Multimodal Retrieval

Universal multimodal retrieval typically encodes a query once and ranks independently indexed items by embedding similarity. This design supports efficient search, but leaves the query representation unchanged even when retrieved items could help clarify the information need. We introduce the AutoRegressive Retriever (ARR), a multimodal retrieval model that learns both to select informative items and to use their content to refine subsequent retrieval. ARR alternates between retrieving an item and updating the query embedding, then uses the final embedding to rank the collection. Supervised fine-tuning teaches the encoder to use feedback through stepwise contrastive supervision. Reinforcement learning treats feedback items as actions and optimizes their selection using the final reciprocal rank of a relevant item. A query-side adapter enables this optimization against a fixed item index. ARR demonstrates strong retrieval performance on both in-domain and zero-shot benchmarks, outperforming the compared baselines on average. Further analyses show that feedback improves retrieval at inference time and that training with feedback also improves the initial query embedding, before any item is observed.

Jianfei Zhao, Yifan Wang, Feng Zhang +7

School of Computer Science and Technology, Beijing Institute of Technology · Zhongguancun Academy · Xiaohongshu +2

Information Retrieval Cross-Modal Retrieval Multimodal Embedding RL for IR

Oct 4, 2026

Enhancing Long-Video VLM Embeddings with Query-Aware Streaming Latent Reasoning

Long-video embedding requires capturing sparse query-relevant evidence under a limited visual-token budget. Uniform sampling can miss brief events in videos spanning minutes or hours, whereas encoding more frames in a single context increases memory and computation. We introduce Query-Aware Streaming Latent Reasoning (QASLR), a post-training framework that accumulates evidence across clips while keeping the embedding size fixed. QASLR selects a bounded set of frames, restores their temporal order, and processes them clip by clip with a vision-language backbone. A compact set of persistent think tokens cross-attends to each clip's features, while an embed token reads out a normalized representation after every update. This design integrates evidence across multiple backbone calls without requiring all selected frames to share a single context. Training combines final contrastive learning, step-wise contrastive supervision, and final-embedding self-distillation. Intermediate supervision trains partial-video readouts for retrieval, while self-distillation regularizes them toward the final representation. Query-aware selection produces query-conditioned representations for candidate-set scoring and reranking, whereas query-independent selection enables reusable corpus indexing. Under the full training recipe, HourVideo retrieval Hit@1 increases from 54.2 to 70.7 and from 57.4 to 72.8 for 2B and 8B Qwen3-VL-Embedding backbones, respectively. Gains extend to the evaluated moment-retrieval and video-QA tasks, and the streaming head transfers to a second Qwen-family embedding backbone. These results support streaming latent aggregation as an effective approach to integrating long-video evidence into fixed-dimensional representations.

Haozhe Chi, Song Jin, Yang Jin +1

Peking University · Renmin University of China

Multimodal Embedding Vision-Language Retrieval Video Retrieval Streaming Video Understanding Query-Aware Video Frame Selection

Oct 2, 2026

World Embedding Benchmark

Physical fidelity has received increasing attention in world models and video generation, yet how video representations encode physical information remains less understood. We introduce the World Embedding Benchmark, comprising 8,000 controlled simulation cases from 80 families spanning fluid mechanics, solid mechanics, dynamics, and optics & electromagnetism. Each case pairs a rendered video with simulation-derived physical annotations, supporting three complementary tasks: text-video retrieval, physical-property regression, and multiple-choice video-description pair classification. We use these tasks to distinguish cross-modal physical alignment from the recoverability of quantitative physical information. Evaluated pre-trained omnimodal embedding models show weak retrieval and near-chance within-family pair classification, while lightweight probes recover useful physical information from frozen video embeddings. Continual contrastive training with physics-specific video-text pairs improves retrieval and pair classification but degrades physical-property regression, revealing a trade-off between alignment and quantitative information recoverability. Finally, we use the embeddings to retrieve reference videos for retrieval-augmented generation with MiniMax-H3. Retrieved references improve the physical fidelity of generated videos, with stronger retrieval models yielding larger gains in our experiments. Together, these findings highlight the need to evaluate physical alignment and property recoverability jointly, and demonstrate the utility of physical representations for improving video generation.

Yiqi Liu, Ruifeng Yuan, Yang Wang +7

World-Embedding Team

Physical Reasoning in Video Video Representation Learning Multimodal Embedding Video Generation Evaluation

Oct 1, 2026

Omni-Embed-Mini: Binding Modalities Without Forgetting via Dense Distillation

Extending a text embedding model to new modalities typically degrades text retrieval quality, and existing omni-modal embedders compensate with multi-billion parameters. We present Omni-Embed-Mini, a 0.9B-parameter model that maps text, speech, audio, images, video, and visually-rich documents into a single shared cosine space without updating any text-side parameter. Our key insight is that the teacher signal requires no separate embedding model: each media sample is paired with a dense cascaded caption, and the teacher target is simply the frozen backbone's own embedding of that caption. Because teacher and student share the same backbone weights, they inhabit byte-identical geometry, and lightweight projectors plus phased LoRA adapters on the modality encoders suffice for alignment. Training combines a Matryoshka SigLIP contrastive loss with an online hybrid hard-negative miner whose negatives sharpen as the encoder improves. The recipe carries over to a 2.3B variant by swapping in a native vision-language backbone. Omni-Embed-Mini-0.9B keeps its text weights bit-identical to the backbone, so training cannot regress text retrieval (49.57 nDCG@10 on MTEB-v2 BEIR-8), while extending it to five additional modalities, and is ~2.7x to 9.5x smaller than every open omni embedder we compare against. The 2.3B variant is competitive with the closed gemini-embedding-2, edging ahead of it on the overall-modality average. Models, code, data and evaluation harness are on our project page: https://omniembed.cvmbzuai.com

Mohammed Irfan Kurpath, Jaseel Muhammad Kaithakkodan, Sahal Shaji Mullappilly +2

Mohamed Bin Zayed University of Artificial Intelligence (MBZUAI)

Cross-Modal Knowledge Distillation Multimodal Embedding Cross-Modal Retrieval Multimodal Contrastive Learning

Sep 30, 2026

Mutual Equilibrium: Multimodal Representation Learning through Reciprocal Feedback

This work proposes a mutual feedback architecture, MEQ, that refines the two inputs, of possibly different modalities, into a pair of coupled embeddings such that each embedding reflects the information of the other. The core idea is to incorporate continuous interchange of information between the two inputs. This idea leads to a mutual feedback architecture consisting of two components whose outputs are fed back into the other. The final output of this model is defined as the fixed point of this interaction. We provide theoretical analysis that offers interpretation of this model as well as design choices to prevent failure cases. We show the benefits of MEQ through classification and visual grounding tasks spanning various datasets. Quantitatively, our model outperforms or shows competitive performance on concatenation-based multimodal classification problems. Qualitatively, the proposed interactive mechanism allows the model to progressively refine the visual grounding when paired with complementary modality, thus demonstrating the power of mutual feedback under such settings.

Ho-min Park, Byungkon Kang

Data Science Center, Texas Children’s Hospital Baylor College of Medicine Houston, Texas 77030, USA · Department of Computer Science SUNY Korea Incheon, Republic of Korea

Multimodal Grounding Multimodal Embedding Multimodal Classification Multimodal Learning

Sep 30, 2026

GATE-ST: Gene-Aware Text-image Encoder for Spatial Transcriptomics

Spatial transcriptomics enables spatially resolved gene expression analysis from slide-level images while preserving morphological features, providing valuable information for studying disease mechanisms and developing treatments. However, spatial gene expression profiling typically requires expensive and time-consuming tests. While existing image-based prediction optimizations mostly revolve around including positional embeddings and further image-based changes, text-based optimizations remain relatively unexplored. We present GATE-ST, which incorporates text-based inputs into image-based spatial gene expression predictions. With this approach, generated text descriptions of genes are utilized to better spatial transcriptomics prediction results. Gene summaries are put through a text encoder, generating embeddings that integrate with image embeddings through cross-attention layers to align with morphological features. We demonstrate the effectiveness of such text inputs by benchmarking performance against random gene embeddings and multiple other image-text fusion architectures, and show that GATE-ST outperforms these alternatives. Our results demonstrate the effectiveness of GATE-ST in pathology imaging, which may greatly reduce the time and cost of accurate spatial transcriptomic predictions, proving the potential of text-guided spatial gene expression prediction.

Lucas Ni, Jian Luo, Wentao Huang +1

Ridge High School Basking Ridge, United States · Stony Brook University Stony Brook, United States

Spatial Transcriptomics Multimodal Embedding Spatial Gene Expression Prediction Gene Expression Prediction from Histopathology Images Whole-Slide Image Analysis

Sep 29, 2026

ResComEmb: Effective and Efficient Multimodal Embedding via Residual Homogeneity Compression

Multimodal large language models (MLLMs) have shown strong potential for universal multimodal representation learning. However, existing methods either compress each input into a single vector, limiting fine-grained expressiveness, or retain long sequences of visual-token vectors, incurring substantial storage and interaction costs. To resolve this trade-off, we propose ResComEmb, a trainable framework for effective and efficient universal multi-vector multimodal embedding. ResComEmb first encodes each input at native dynamic resolution into ordered global, intermediate, and fine-grained views. After MLLM contextualization and embedding projection, a trainable Residual Homogeneity Compression (RHC) module reduces within-granularity redundancy and cross-granularity repetition under explicit visual token budgets. Then, ResComEmb introduces a length-adaptive Bidirectional Late-Interaction Matching mechanism for robust query-document scoring, which averages the strongest token-level matches in each direction and combines the two scores using a weight based on how many valid tokens each side has. Extensive experiments on MMEB, ViDoRe V1, and ViDoRe V2 show that ResComEmb produces higher-quality universal multimodal embeddings than VLM2Vec-V2, and outperforms ColQwen2.5 in visual document retrieval using only 37.5% of its full visual token budget, demonstrating a favorable effectiveness-efficiency trade-off.

Zijing Cai, Yuzhe Wang, Jingxian Zhu +2

University of Science and Technology of China · Hefei University of Technology · National University of Singapore

Multimodal IR Multimodal Embedding Multi-Vector Retrieval Multimodal Token Compression Visual Document Retrieval

Sep 27, 2026

Binding Multiple Modalities via Multimodal Wasserstein Barycenter

Multimodal learning beyond two modalities commonly leverages a specific modality (e.g., text) to bind other modalities. However, how to establish a more balanced representation space that approximates shared semantics while respecting the holistic geometry of

n

-modal data remains challenging. In this work, we present BaryBind, which aims to transport the specific modality towards the Wasserstein barycenter (WB) optimized across all modalities and introduces a volumetric alignment objective to establish a unified semantic space around the WB embedding. Specifically, we project specific modalities to the WB, which minimizes the average Wasserstein distances to multimodal distributions and serves as the anchor for subsequent alignment. We then construct a barycenter simplex, whose volume is taken as a similarity metric for global alignment centered at the WB. Experiments show that BaryBind achieves competitive performance in text-video-audio retrieval, classification, videoQA, and cross-modal generation tasks, along with robustness under modality absence and scalability to more than three modalities. Code is released at https://github.com/xl-tang3/BaryBind.

Xiaole Tang, Jiayi Xu, Xiang Gu +2

Xi’an Jiaotong University

Cross-Modal Alignment Multimodal Embedding Wasserstein Barycenters

Sep 27, 2026

Learning Multimodal Embeddings with Evidence-Aligned Readout

Multimodal large language models can expose task-relevant evidence through generation, but producing useful evidence does not by itself determine how it enters a retrieval embedding. We study whether the semantic organization of that evidence can also specify where representations are read. To address this question, we introduce EviAlign, which couples Semantic Evidence Generation with Boundary Readout in a shared multimodal large language model. It organizes evidence into five semantic units, reads the contextualized state at each unit boundary, and aggregates these states into a single normalized embedding. Generation and contrastive retrieval objectives jointly train this shared structure. With the same trailing readout, semantic evidence and free-form CoT yield nearly identical retrieval performance, suggesting that evidence organization alone does not explain the full gain. A controlled

2\times3

study compares consistent and permuted evidence organization across three readout strategies, using training targets with matched evidence spans. With five readout states and the same mean pooling, the advantage of consistent semantic organization grows from 0.65 points at length-based training positions to 2.39 at evidence boundaries, yielding a 1.74-point co-design interaction. Across 12 MMEB retrieval tasks, EviAlign achieves 76.9 average Recall@1 with 500K training pairs while retaining single-vector indexing and scoring.

Zirong Chen, Fuda Ye, Enjun Du +8

The Hong Kong University of Science and Technology (Guangzhou) · Tencent Yuanbao · Tsinghua University +3

Multimodal Large Language Models Multimodal Embedding Cross-Modal Retrieval Multimodal Contrastive Learning

Sep 27, 2026

VaME: Exploring Variational Latent Reasoning for Multimodal Embeddings

Universal multimodal retrieval requires compact embeddings that preserve task-relevant semantic information across diverse modalities. Prior works have incorporated latent reasoning into multimodal embedding learning to refine this information before embedding extraction. However, most existing approaches remain confined to deterministic latent paths, without exploring alternative trajectories to discover better embeddings. Thus, we propose VaME (Variational Multimodal Embeddings), a framework that models latent reasoning as a learnable distribution over trajectories. Specifically, we first introduce Variational Latent Reasoning (VLR) to enable autoregressive exploration in latent space, guided by answer reconstruction through a lightweight decoder. Meanwhile, we augment the original embedding-token readout with a latent-fused embedding to facilitate exploration during subsequent reinforcement learning. Finally, we optimize latent reasoning over stochastic variational trajectories through reinforcement learning, using Semantic Decoding Reward (SDR) to favor semantically meaningful trajectories with interpretable decoded outcomes. On the 78-task MMEB-V2 benchmark, spanning image, video, and visual-document retrieval, VaME outperforms most explicit CoT-based models and all latent-reasoning baselines. VaME also demonstrates robust performance on reasoning-intensive benchmarks such as MRMR, with substantial gains after reinforcement learning. Importantly, VaME achieves these gains with at least a 4.25x inference speedup over the deterministic latent autoregressive baselines. The code will be made publicly available.

Peixi Wu, Mingzhou Jiang, Feipeng Ma +11

University of Science and Technology of China · Tsinghua University · Kuaishou +1

Multimodal IR Efficient Multimodal Inference Multimodal Reasoning Multimodal Embedding Latent Reasoning

Sep 25, 2026

BAT-CLIP: Trimodal Alignment of Brain, Audio and Text

Decoding and interpreting naturalistic speech from the brain increasingly relies on alignment to pretrained speech and language representation spaces. However, current CLIP-style brain-speech alignment ground neural activity to a single anchor modality-audio or text-despite the brain's inherently multimodal speech processing. This induces a trade-off: audio anchoring preserves temporal structure but weakens linguistic separability, while text anchoring captures semantics yet discards acoustic detail. We propose BAT-CLIP, the first CLIP-style trimodal alignment framework for iEEG that jointly aligns neural embeddings to both pretrained audio and text anchors in a shared, frozen audio-text manifold. On the naturalistic Podcast benchmark, BAT-CLIP yields more robust representations than bimodal CLIP baselines. We also highlight the importance of using self-supervised foundation models for CLIP training.

Suhyun Kim, Jinmo Han, Danny Dongyeop Han +9

Yonsei University, Seoul, Republic of Korea · Seoul National University, Seoul, Republic of Korea · Dartmouth College, Hanover, NH, USA

Cross-Modal Alignment Multimodal Embedding Neural Decoding

Sep 23, 2026

LAYERSCOPE: A Layerwise Characterization of Video and Multimodal Learned Representations

We propose LAYERSCOPE, a label-free, layerwise framework that aims to characterize a model's learned representations in video and multimodal settings. Evaluating downstream performance using representations from final or intermediate layers typically requires large amounts of labeled data, repeated task-specific evaluations, and substantial computation. To address these limitations, LAYERSCOPE uses local, global, distributional, and correspondence-based geometric metrics to compare layerwise representation structure within and across models without requiring task-specific labels. We evaluate seven architecturally diverse models across video and multimodal classification, clustering, and text-to-video retrieval tasks from MVEB/MVEB+. We find that intermediate-layer representations can outperform final-layer and model-default outputs. We also find that no single geometric metric consistently predicts downstream performance, but note that distinct layerwise geometric signatures emerge across model families. LID shows task-dependent relationships with performance, while RankMe provides the strongest measure for classification and clustering, but is not a universal layer selector. We also find that pairing-aware metrics explain retrieval better than distributional distances alone. LAYERSCOPE therefore offers a framework for comparing representations across models and layers, enabling a more systematic evaluation in video and multimodal settings.

Sandra Arcos-Holzinger, Debashish Chakraborty, Rohita Mocharla +7

Johns Hopkins University · University of Melbourne · Human Language Technology Center of Excellence +2

Representation Geometry Representation Learning Multimodal Embedding Multimodal Model Evaluation

Sep 21, 2026

Ovis-Embedding: Pushing the Frontiers of Universal Omni-Modal Embeddings

In this report, we introduce Ovis-Embedding, a state-of-the-art omni-modal embedding family built on native integration of text, image, video, and audio. Instead of assembling separate modality towers, Ovis-Embedding uses a shared multimodal backbone to encode different modalities in a common representation space. Specifically, we make three key advances: (1) native omni-modal initialization: we adopt a pretrained Qwen-omni model as the embedding backbone and adapt it through contrastive training with low-rank initialization; (2) data-centric omni-modal training: we construct a broad, high-quality corpus spanning text, images, video, audio, and interleaved multimodal data. To improve data efficiency, we introduce homogeneous-source sampling to form task-consistent batches with informative in-batch negatives; and (3) embedding-specific training and inference optimization: we use focal loss to emphasize hard examples and similarity-based Embedding Distillation to transfer fine-grained similarity structure from complementary experts. At inference time, low-rank feature decomposition enables compact embeddings with flexible dimensionality and minimal performance loss. Empirical evaluations show that the Ovis-Embedding family achieves state-of-the-art performance on MMEB-v3, MMEB-v2, MVEB, MAEB, and RTEB, demonstrating its effectiveness across text, image, video, and audio modalities. These results highlight the potential of unified omni-modal training to overcome modality fragmentation and advance universal embedding models for any-to-any retrieval.

Ovis-Embedding Team

Alibaba Token Hub, Alibaba Group

Unified Multimodal Models Multimodal Embedding Cross-Modal Retrieval Multimodal Contrastive Learning

Sep 16, 2026

SCOUT: Sim-to-Real Text-Based Person Retrieval by Embedding-Space Prediction over Frozen Video Features

Text-based person retrieval under a sim-to-real gap (synthetic training data, a real-image gallery) is usually tackled with costly fine-tuned cross-encoders. We ask whether a frozen-encoder system can compete. We present SCOUT, which casts cross-modal retrieval as prediction in embedding space. A trainable predictor maps the patch tokens of a frozen video encoder into the embedding space of a frozen text encoder under a bidirectional InfoNCE objective, and no encoder is fine-tuned in the base model. The video encoder is V-JEPA, the text encoder is EmbeddingGemma, and the predictor is initialized from a Qwen3.5-0.8B decoder. We make three findings. First, the best frozen text encoder is simply the one whose geometry best matches the video features. A training-free alignment score ranks three candidate text encoders in the same order as their retrieval accuracy on our held-out split (Spearman

ρ= 1.0

); a fourth, LLM-based encoder shows the rule is metric-dependent, holding for a neighborhood-overlap score (

ρ= 0.8

) but not for a linear probe (

ρ= -0.2

). Second, two precision-targeted levers, parameter-efficient ExPLoRA adaptation of the video encoder and a training-free attribute-decomposed reranker built on a vision-language model, improve the top-rank precision that otherwise limits the frozen system, adding 2.2 points of leaderboard R@1. Third, a local-versus-public calibration study explains which interventions transfer to the real domain. On AI City Challenge 2026 Track 4 the full retrieve-fuse-rerank system reaches 84.25 mAP@10 on the final leaderboard, while a single frozen model submitted alone reaches 60.63. Our trained components cost about 95 GPU-hours. CMP, the dataset authors' fine-tuned cross-encoder that trains for sixteen GPU-days, is one fusion member of the full system, not an alternative. Code and annotations: https://github.com/abtraore/SCOUT-ECCV

Abdarahmane Traoré, Andy Couturier, Éric Hervet

Embia, Computer Science Department, Faculty of Science, Université de Moncton, Moncton, NB, Canada

Text-Based Person Search Synthetic-to-Real Domain Adaptation Multimodal Embedding Cross-Modal Retrieval Vision-Language Alignment

Sep 15, 2026

Multi-modal Knowledge Preserving Adapter for Embedding Backward Compatibility

Upgrading embedding models typically requires expensive database re-indexing, as new query embeddings are incompatible with existing database embeddings. While Backward Compatible Training (BCT) mitigates this by enforcing compatibility during training, existing approaches often require updating the backbone model. This is impractical because of significant training cost, the risk of performance regression, and limited access to proprietary model weights. We introduce Multi-modal Knowledge Preserving Adapter (MKP-Adapter), the first adapter-only BCT approach for Multi-modal Large Language Models (MLLMs) that requires no backbone updates. We identified that the primary challenge in adapter-only BCT is preserving the knowledge of the new embeddings while enforcing backward compatibility. Hence, we propose a multi-level preservation loss that maintains the geometric structure of the embedding spaces throughout BCT. Furthermore, a focal re-weighting strategy is integrated to prioritize learning from challenging samples. Experiments demonstrate that our method achieves strong backward compatibility across diverse multi-modal benchmarks (image, text, visual document, and video retrieval tasks) and model types. Notably, MKP-Adapter is trained solely on pre-extracted embeddings and requires only negligible additional latency relative to the original backbone forward pass, highlighting its efficiency.

Jaeseok Byun, Gukyeong Kwon, Han-Kai Hsu +4

1Seoul National University, South Korea · 2Amazon AGI, USA

Multimodal Large Language Models Adapter Tuning Multimodal Embedding

Sep 14, 2026

Hypergraph-Regularized Gramian Volumes for Multimodal Retrieval

Volume-based multimodal retrieval jointly scores a text query with a candidate's video, audio, and subtitle embeddings. While this approach captures higher-order within-candidate alignment, the score remains candidate-local, and semantically related training samples primarily serve as contrastive negatives. This work introduces Hypergraph-Regularized Gramian Volumes (HyVol), a training-time module that incorporates these semantic relations prior to evaluating the original volume loss. Document hyperedges connect the observed modalities of each candidate, whereas semantic hyperedges link candidates whose detached captions are mutual top-k neighbors. A shallow gated hyper-graph network applies residual corrections to the modality embeddings. Presence masks exclude unavailable streams from message passing, and identity padding preserves the determinant of the observed Gram submatrix without feature imputation. As refinement operates on embeddings rather than scores, the same construction applies to both Gram and HyperGram. We remove the hypergraph after training, leaving the backbone-only architecture, original scoring function, and retrieval cost unchanged. We train both backbones on a 150K-clip subset of VAST-27M and evaluate zero-shot performance on six benchmarks. Under the paired protocol, HyVol improves R@1 across all five retrieval benchmarks, with video-to-text gains reaching +8.3 on MSR-VTT and +7.6 on VATEX. Under missing-modality masking, the V2T margin remains positive in all experimental settings, although the T2V margin becomes slightly negative in four.

Anindya Nag, Ambuj Mehrish, Sebastiano Vascon

Department of Environmental Science, Informatics and Statistics Ca’ Foscari University of Venice, Italy

Multimodal IR Cross-Modal Alignment Multimodal Embedding Hypergraph Neural Networks

Sep 14, 2026

Reason What Matters: Retrieval-Grounded Reasoning for Universal Multimodal Embeddings

Universal multimodal embedding (UME) maps multimodal inputs into a shared embedding space for diverse retrieval tasks. Recent methods improve embeddings through Chain-of-Thought (CoT) reasoning optimized with GRPO using retrieval rewards. However, existing methods overlook the mismatch bettween candidate-aware retrieval supervision and input-only CoT generation: (1)trajectory-level rewards convey retrieval outcomes without explicitly identifying the input-supported evidence that distinguishes the positive from hard negatives; (2) input-only generation cannot directly assess whether further reasoning improves retrieval, potentially producing redundant CoTs with substantial latency. To bridge this gap, we propose Reason What Matters (ReWAM), a retrieval-grounded framework that aligns candidate-aware supervision with input-only generation. Specifically, we introduce Retrieval-Aware Self-Distillation (RASD), which extracts privileged guidance from input-supported facts and evidence distinguishing the positive from hard negatives. Conditioned on this guidance, an on-policy self-teacher provides token-level feedback to refine credit assignment, directing policy updates toward retrieval-relevant reasoning grounded in the input. We further propose Retrieval-Adaptive Inference (RAI), which learns a retrieval-aware stopping criterion from prefix-level retrieval feedback. It stops redundant reasoning without candidate access and uses speculative decoding to further reduce CoT latency. Extensive experiments on MMEB-V2 and MRMR demonstrate that ReWAM achieves state-of-the-art retrieval performance while delivering up to 5x the inference throughput of competitive explicit-CoT UME methods. ReWAM thus enables high-quality retrieval through efficient input-only reasoning, making explicit CoT practical for corpus-scale multimodal retrieval. The code will be publicly available.

Mingzhou Jiang, Peixi Wu, Hang Cheng +7

Tsinghua Shenzhen International Graduate School, Tsinghua University · School of Artificial Intelligence and Data Science, USTC · Kuaishou Technology +1

Multimodal IR Efficient Multimodal Inference RL for Language Model Reasoning Multimodal Embedding Multimodal CoT Reasoning