Multimodal Foundation Models

Follow this topic Ask about this topic

Momentum

13 papers in the last four weeks, up 117% on the four weeks before. 0.1% of all new papers.

Jul 13Week of Sep 28

Latest papers 303

May 18, 2026cs.CV

Vision-OPD: Learning to See Fine Details for Multimodal LLMs via On-Policy Self-Distillation

Multimodal Large Language Models (MLLMs) still struggle with fine-grained visual understanding, where answers often depend on small but decisive evidence in the full image. We observe a regional-to-global perception gap: the same MLLM answers fine-grained questions more accurately when conditioned on evidence-centered crops than on the corresponding full images, suggesting that many failures stem from difficulty to focus on relevant evidence rather than insufficient local recognition ability. Motivated by this observation, we propose Vision-OPD (Vision On-Policy Distillation), a regional-to-global self-distillation framework that transfers the model's own privileged regional perception to its full-image policy. Vision-OPD instantiates two conditional policies from the same MLLM: a crop-conditioned teacher and a full-image-conditioned student. The student generates on-policy rollouts, and Vision-OPD minimizes token-level divergence between the teacher and student next-token distributions along these rollouts. This enables the model to internalize the benefit of visual zooming without external teacher models, ground-truth labels, reward verifiers, or inference-time tool use. Experiments on multiple fine-grained visual understanding benchmarks show that Vision-OPD models achieve competitive or superior performance against much larger open-source, closed-source, and "Thinking-with-Images" agentic models. The code is available at https://github.com/VisionOPD/Vision-OPD

Qianhao Yuan, Jie Lou, Xing Yu +4

Chinese Information Processing Laboratory, Institute of Software, Chinese Academy of Sciences · University of Chinese Academy of Sciences · Xiaohongshu Inc.

Multimodal Large Language Models On-Policy Distillation Multimodal Foundation Models Knowledge Distillation Fine-Grained Visual Perception Multimodal Understanding

May 18, 2026cs.CV

Lance: Unified Multimodal Modeling by Multi-Task Synergy

We present Lance, a lightweight native unified model supporting multimodal understanding, generation, and editing for both images and videos. Rather than relying on model capacity scaling or text-image-dominant designs, Lance explores a practical paradigm for unified multimodal modeling via collaborative multi-task training. It is grounded in two core principles: unified context modeling and decoupled capability pathways. Specifically, Lance is trained from scratch and employs a dual-stream mixture-of-experts architecture on shared interleaved multimodal sequences, enabling joint context learning while decoupling the pathways for understanding and generation. We further introduce modality-aware rotary positional encoding to mitigate interference among heterogeneous visual tokens and boost cross-task alignment. During training, Lance adopts a staged multi-task training paradigm with capability-oriented objectives and adaptive data scheduling to strengthen both semantic comprehension and visual generation performance. Experimental results demonstrate that Lance substantially outperforms existing open-source unified models in image and video generation, while retaining strong multimodal understanding capabilities. The homepage is available at https://lance-project.github.io.

Fengyi Fu, Mengqi Huang, Shaojin Wu +10

Intelligent Creation Lab, ByteDance

Unified Multimodal Models Multi-Task Learning Multimodal Generation Multimodal Foundation Models Multimodal Learning Multimodal Understanding

May 18, 2026cs.CV

RAVE: Re-Allocating Visual Attention in Large Multimodal Models

Large multimodal models (LMMs) inherit the self-attention mechanism of pretrained language backbones, yet standard attention can exhibit suboptimal allocation, including cross-modal misallocation between textual and visual evidence and intra-visual imbalance among visual tokens. We propose RAVE (Re-Allocating Visual Attention), a lightweight pair-gating mechanism that adds a learned query-key bias to pre-softmax attention scores over visual keys, derived from pre-RoPE query and key features. RAVE requires no architectural modification to the backbone and can be trained end-to-end with the rest of the model. Across a suite of multimodal benchmarks, RAVE improves over standard attention by an average of 3 points, with the largest gains on perception-intensive tasks -- including multilingual OCR, chart understanding, document VQA, and scene text VQA -- where accurate visual grounding is critical.

Xi Leng, Xinhong Ma, Ziqiang Dong +4

Qwen Business Unit of Alibaba · The Chinese University of Hong Kong, Shenzhen · Beijing Institute of Technology

Visual Attention Self-Attention Multimodal Large Language Models Cross-Modal Attention Attention Mechanisms Multimodal Foundation Models+2

May 18, 2026cs.CV

A More Word-like Image Tokenization for MLLMs

Modern multimodal large language models (MLLMs) typically keep the language model fixed and train a visual projector that maps the pixels into a sequence of tokens in its embedding space, so that images can be presented in essentially the same form as text. However, the language model has been optimized to operate on discrete, semantically meaningful tokens, while prevailing visual projectors transform an image into a long stream of continuous and highly correlated embeddings. This causes the visual tokens to behave differently from the word-like units that LLMs are originally trained to understand. We propose a novel Disentangled Visual Tokenization (DiVT) that clusters patch embeddings into coherent semantic units, so each token corresponds to a distinct visual concept instead of a rigid grid cell. DiVT further adapts its token budget to image complexity, providing an explicit accuracy-compute trade-off modifying neither the vision encoder nor the language model. Across diverse multimodal benchmarks, DiVT matches or surpasses baselines with significantly fewer visual tokens, demonstrating robustness under limited token budgets, significantly reducing memory cost and latency while making visual inputs more compatible with LLMs. Our code is available at https://github.com/snuviplab/DiVT.

Hyun Lee, Hyemin Jeong, Yejin Kim +4

Seoul National University · Ewha Womans University

Image Tokenization Efficient VLM Inference Visual Tokenization Multimodal Large Language Models Multimodal Foundation Models Visual Token Compression+1

May 18, 2026cs.CV

LatentUMM: Dual Latent Alignment for Unified Multimodal Models

Unified multimodal models (UMMs) achieve strong performance in both understanding and generation by learning a shared latent space, yet they often exhibit functional inconsistency between these two capabilities. We observe that this issue does not stem from a lack of shared representations, but from the absence of explicit alignment between the transformations that map into and out of the latent space. As a result, generation and re-encoding can follow inconsistent trajectories, leading to semantic drift under modality transitions. In this work, we propose LatentUMM, a framework that constructs an enhanced shared latent space to explicitly align these transformations and improve cross-modal consistency. LatentUMM consists of two stages. First, dual latent alignment enforces consistency at both the modality and capacity levels: cross-modal alignment uses a stronger embedding model to impose structured cross-modal semantics, while dual capacity alignment enforces bidirectional consistency under generation and re-encoding. Second, latent dynamics stabilization improves robustness via stochastic latent rollouts and preference optimization, favoring trajectories that better preserve semantic consistency. Experiments show that LatentUMM consistently improves multimodal consistency across diverse architectures. Code is available at: https://github.com/AIFrontierLab/TorchUMM/tree/main/src/umm/post_training/LatentUMM.

Yinyi Luo, Wenwen Wang, Hayes Bai +2

Carnegie Mellon University · William & Mary

Cross-Modal Learning Unified Multimodal Models Cross-Modal Alignment Multimodal Foundation Models Latent Space Alignment Multimodal Understanding

May 16, 2026cs.CL

Multilingual and Multimodal LLMs in the Wild: Building for Low-Resource Languages

Multimodal LLMs are evolving from vision-language to tri-modality that see, hear, and read, yet pipelines and benchmarks remain English-centric and compute-heavy. The tutorial offers an overview of this emerging research area for multilingual multimodality across text, speech, and vision under limited data/compute budgets, synthesizing foundations, recent multilingual models (PALO, Maya), speech-text LLMs. We cover low-cost data creation/curation; adapter stacks for tri-modal alignment; culture-aware evaluation beyond English and hands on resources for fine-tuning a compact multilingual VLM and wiring a speech->text->LLM pipeline. The content will be delivered as an interactive half-day tutorial, designed for researchers and practitioners working on multilingual, multimodal AI in low-resource language settings.

Firoj Alam, Shammur Absar Chowdhury, Enamul Hoque Prince

Qatar Computing Research Institute, HBKU, Qatar · York University, Canada

Cross-Modal Alignment Multimodal Large Language Models Multimodal Foundation Models Low-Resource Language Processing Multimodal Learning Multimodal Understanding

May 15, 2026cs.AI

TTE-Flash: Accelerating Reasoning-based Multimodal Representations via Think-Then-Embed Tokens

Recent research has demonstrated that Universal Multimodal Embedding (UME) benefits significantly from Chain-of-Thought (CoT) reasoning. In this paradigm, a generative model produces explicit reasoning traces for a multimodal query, with the final representation extracted from an <eos> embedding token attending to both the query and the reasoning. Despite its effectiveness, the computational overhead of generating explicit CoT traces is often prohibitive. In this work, we propose replacing explicit CoT with latent think tokens, which are interpreted as latent variables that can produce explicit CoT traces as observed variables. By optimizing think tokens using CoT generation loss and subsequent embedding tokens using contrastive loss, we produce high-performance, reasoning-aware representations at a constant inference cost. Our study investigates two key architectural designs: 1) how think and embeddings tokens should be extracted from the same LLM backbone. 2) how the tokens should be trained as two dependent tasks. We introduce TTE-Flash-2B, a reasoning-aware multimodal representation model that outperforms its explicit-CoT counterpart on the MMEB-v2 benchmark, while producing latent think tokens that are interpretable both textually and visually. Furthermore, zero-shot evaluation across 15 video datasets reveals scaling behavior as the number of think tokens increases, and motivating a pilot study of adaptive think budget allocation based on task requirements.

Jianpeng Cheng, Xian Wu, Jiangfan Zhang +10

1Meta AI

Efficient Multimodal Inference Multimodal Embedding Multimodal CoT Reasoning Multimodal Foundation Models Multimodal Understanding

May 15, 2026cs.CV

Second-Order Multi-Level Variance Correction for Modality Competition in Multimodal Models

Autoregressive next-token training offers a unified formulation for image generation and text understanding, but it also creates strong modality competition that destabilizes optimization and limits large-batch scaling. We show that first-order optimizers such as AdamW are vulnerable to cross-modality gradient heterogeneity, while second-order preconditioning, particularly SOAP, provides a more stable basis for multimodal alignment. Building on this insight, we propose \emph{ML-FOP-SOAP}, a second-order optimization framework with Multi-Level Variance Correction. Our Fisher-Orthogonal Projection suppresses variance-induced modality conflicts, reducing the trade-off between visual generation and textual understanding. To make this practical under large gradient accumulation, we introduce a hierarchical folding strategy that captures fine-grained variance with low micro-step overhead. Experiments on Janus and Emu3 show consistent gains across both modalities and stable training at batch size 8192. Compared with AdamW, our method improves sample efficiency by up to

1.4\times

and accelerates wall-clock training by up to

1.5\times

, offering a robust optimizer for scaling multimodal foundation models.

Yishun Lu, Wes Armour

University of Oxford, Oxford, United Kingdom

Second-Order Optimization Multimodal Optimization Multimodal Foundation Models Multimodal Learning Multimodal Understanding

May 15, 2026cs.CR

DarkLLM: Learning Language-Driven Adversarial Attacks with Large Language Models

While vision and multimodal foundation models underpin critical tasks from perception to complex reasoning, they remain highly vulnerable to adversarial attacks. However, traditional adversarial attacks are typically limited to single, predefined objectives, tightly coupling each attack to a specific model or task, which restricts their scalability and flexibility in real-world scenarios. In this work, we present DarkLLM, a novel attack framework that trains an LLM to translate natural-language attack instructions into latent attack vectors, which are then decoded into visual adversarial perturbations. By leveraging natural-language instruction tuning, DarkLLM not only unifies targeted, untargeted, segmentation, and multi-model attacks within a single framework, but also achieves flexible and controllable adversarial generation, enabling each instruction to produce a perturbation that induces desired behaviors across heterogeneous models. Through extensive experiments across 4 tasks, 13 datasets, and 15 models, we demonstrate that DarkLLM with only 1B parameters can follow attacker instructions and generate highly effective attacks against CLIP, SAM, and frontier LLMs, revealing a systemic vulnerability in modern foundation models.

Ye Sun, Xin Wang, Jiaming Zhang +7

Fudan University · Nanyang Technological University · Tongji University

Adversarial Attacks on VLMs Adversarial Perturbations Multimodal Foundation Models

May 15, 2026cs.CV

Unlocking Dense Metric Depth Estimation in VLMs

Vision-Language Models (VLMs) excel at 2D tasks such as grounding and captioning, yet remain limited in 3D understanding. A key limitation is their text-only supervision paradigm, which under-constrains fine-grained visual perception and prevents the recovery of dense geometry. Prior methods either distill geometry from external vision models, introducing error accumulation, or enable direct prediction with inefficient per-pixel query or coarse token-level outputs. In this paper, we propose DepthVLM, a simple yet effective framework that transforms a single VLM into a native dense geometry predictor while preserving its multimodal capability. By attaching a lightweight depth head to the LLM backbone and training under a unified vision-text supervision paradigm with a two-stage schedule, DepthVLM generates full-resolution depth maps alongside language outputs in a single forward pass. We further introduce a unified indoor-outdoor metric depth benchmark in a VLM-compatible format. Experiments show that DepthVLM significantly outperforms existing VLMs with higher inference efficiency, surpasses leading pure vision models, and improves complex 3D spatial reasoning, moving toward a truly unified multimodal foundation model. The project page is available at https://depthvlm.github.io/

Hanxun Yu, Xuan Qu, Yuxin Wang +2

Zhejiang University · Tencent Hunyuan LLM · HKUST +1

Vision-Language Models Metric Depth Estimation Depth Estimation Monocular Depth Estimation Multimodal Foundation Models 3D Scene Understanding

May 15, 2026cs.CV

Reversing the Flow: Generation-to-Understanding Synergy in Large Multimodal Models

The long-standing goal of multimodal AI is to build unified models in which visual understanding and visual generation mutually enhance one another. Despite recent works such as BAGEL, BLIP3o achieves remarkable progress; In practice, however, this unification remains one-directional: understanding routinely guides generation, yet how and why generation can support understanding is rarely investigated. We revisit this asymmetry and propose Generation-to-Understanding (G2U) synergy, where visual generation becomes an explicit intermediate reasoning step. Our framework enables a model to perform controlled generative acts, such as detail enhancement, context expansion or structural visualisation, to produce self-generated visual thoughts, which are then fed back into the model to refine perception without retraining or external tools. Through a comprehensive evaluation on twelve benchmarks, this reversed information flow consistently improves multimodal understanding. We show that generative fidelity bounds perceptual gain and that distinct families of edit prompts govern transfer efficiency. We further analyse whether models can decide what to imagine. While they can produce plausible edits, these self-generated visual thoughts lack stable task alignment, revealing that current large multimodal models fall short of true self-reflection. This work exposes a missing mechanism in unified cognition and suggests that imagination is not the end of understanding but its beginning.

Yujun Tong, Dongliang Chang, Zijin Yin +3

Beijing University of Posts and Telecommunications · Beijing Key Laboratory of Multimodal Data Intelligent Perception and Governance

Unified Multimodal Models Multimodal Reasoning Multimodal Generation Multimodal Foundation Models Multimodal Learning Multimodal Understanding

May 14, 2026cs.RO

Pelican-Unify 1.0: A Unified Embodied Intelligence Model for Understanding, Reasoning, Imagination and Action

We present Pelican-Unify 1.0, the first embodied foundation model trained according to the principle of unification. Pelican-Unify 1.0 uses a single VLM as a unified understanding module, mapping scenes, instructions, visual contexts, and action histories into a shared semantic space. The same VLM also serves as a unified reasoning module, autoregressively producing task-, action-, and future-oriented chains of thought in a single forward pass and projecting the final hidden state into a dense latent variable. A Unified Future Generator (UFG) then conditions on this latent variable and jointly generates future videos and future actions through two modality-specific output heads within the same denoising process. The language, video, and action losses are all backpropagated into the shared representation, enabling the model to jointly optimize understanding, reasoning, imagination, and action during training, rather than training three isolated expert systems. Experiments demonstrate that unification does not imply compromise. With a single checkpoint, Pelican-Unify 1.0 achieves strong performance across all three capabilities: 64.7 on eight VLM benchmarks, the best among comparable-scale models; 66.03 on WorldArena, ranking first; and 93.5 on RoboTwin, the second-best average among compared action methods. These results show that the unified paradigm succeeds in preserving specialist strength while bringing understanding, reasoning, imagination, and action into one model.

Yi Zhang, Yinda Chen, Che Liu +26

Beijing Innovation Center of Humanoid Robotics (X-Humanoid) WFM System Group

Action-Conditioned Video Generation Unified Multimodal Models VLM Reasoning Vision-Language-Action Models Multimodal Foundation Models Reasoning for VLA Models+3

May 14, 2026cs.CV

Do We Really Need External Tools to Mitigate Hallucinations? SIRA: Shared-Prefix Internal Reconstruction of Attribution

Large vision-language models (LVLMs) often hallucinate when language priors dominate weak or ambiguous visual evidence. Existing contrastive decoding methods mitigate this problem by comparing predictions from the original image with those from externally perturbed visual inputs, but such references can introduce off-manifold artifacts and require costly extra forward passes. We propose SIRA, a training-free internal contrastive decoding framework that constructs a counterfactual reference inside the same LVLM by exploiting the staged information flow of multimodal transformers. Instead of removing visual information from the input, SIRA first lets image and text tokens interact through a shared prefix, forming an aligned multimodal state that preserves prompt interpretation, decoding history, positional structure, and early visual grounding. It then forks a counterfactual branch in later transformer layers, where attention to image-token positions is masked. This branch retains the shared multimodal context but lacks continued access to fine-grained visual evidence, yielding a language-prior-dominated internal reference for token-level contrast. During decoding, SIRA suppresses tokens that remain strong without late visual access and favors predictions whose advantage depends on the full visual pathway. Experiments on POPE, CHAIR, and AMBER with Qwen2.5-VL and LLaVA-v1.5 show that SIRA consistently reduces hallucinations while preserving descriptive coverage and incurring lower overhead than two-pass contrastive decoding. SIRA requires no training, external verifier, or perturbed input, and applies to open-weight LVLMs with white-box inference access.

Tian Qin, Junzhe Chen, Yuqing Shi +3

Tsinghua University · The University of Sydney · Stanford University +1

Efficient VLM Inference Hallucination in Language Models VLM Hallucination Large Vision-Language Models Multimodal Foundation Models Contrastive Decoding+2

May 13, 2026cs.CV

Unified Pix Token And Word Token Generative Language Model

Since the emergence of Vision Transformer (ViT), it has been widely used in generative language model and generative visual model. Especially in the current state-of-art open source multimodal models, ViT obtained by CLIP or SigLIP method serves as the vision encoder backbone to help them acquire visual understanding capabilities. But this method leads to limitations in visual understanding for details, such as difficulty in recognizing small text or numbers in images. To address these issues, we propose a new model to unify pix token and word token into the generative language model. The new model also features with each pix of image having its own token embedding, color folding, global conditional attention approximation and image unsupervised pretraining. We conducted image unsupervised pretraining experiments using our new model to explore its potential. The experimental results show that it has good performance even in small model and with limited training data. We believe our model also conforms to the scaling law, as long as model parameters and training data increased, its performance will continue to improve.

Haun Leung, ZiNan Wang

Image Tokenization Self-Supervised Pre-Training Unified Multimodal Models Visual Tokenization Multimodal Foundation Models Vision-Language Pretraining

May 13, 2026cs.AI

Senses Wide Shut: A Representation-Action Gap in Omnimodal LLMs

When an omnimodal large language model accepts a question whose textual premise contradicts what it actually sees or hears, does the failure lie in perception or in action? Recent omnimodal models are positioned as perception-grounded agents that jointly process video, audio, and text, yet a basic form of grounding remains untested: catching a textual claim that conflicts with the model's own sensory input. We introduce IMAVB, a curated 500-clip benchmark of long-form movies with a 2x2 design crossing target modality (vision, audio) and premise condition (standard, misleading), which lets us measure conflict detection separately from ordinary multimodal comprehension. Across eight open-source omnimodal LLMs and Gemini 3.1 Pro, we document a Representation-Action Gap: hidden states reliably encode premise-perception mismatches even when the same models almost never reject the false claim in their outputs. Behaviorally, models fall into two failure modes: under-rejection, in which they answer misleading questions as if the false premise were true; and over-rejection, in which they reject more often but also reject standard questions, sacrificing ordinary comprehension accuracy. The gap is modality-asymmetric (audio grounding underperforms vision) and prompt-resistant across seven variants. As an initial diagnostic intervention, a probe-guided logit adjustment (PGLA) re-injects the encoded mismatch signal into decoding and consistently improves rejection behavior. Together, these results suggest the bottleneck for omnimodal grounding lies in translation, not perception.

Trung Nguyen Quang, Yiming Gao, Fanyi Pu +3

Nanyang Technological University · LMMs-Lab Team · Johns Hopkins University

Multimodal Grounding Multimodal Large Language Models Multimodal Reasoning Multimodal Foundation Models Inference-Time Intervention Multimodal Understanding

May 13, 2026cs.CV

Multilingual OCR-Aware Fine-Tuning and Prompt-Guided Chain-of-Thought Reasoning for Multimodal Large Language Models

Optical character recognition (OCR) and multilingual text understanding remain major failure modes of multimodal large language models (MLLMs), particularly in real-world images containing cluttered layouts, small fonts, blur, occlusion, and complex typography. We present an OCR-aware multilingual multimodal training framework that combines (i) large-scale synthetic OCR-to-translation data generation, (ii) OCR-aware supervised fine-tuning (SFT) with LoRA adaptation, and (iii) structured visual chain-of-thought (CoT) prompting for reasoning under uncertain visual conditions. Using a LLaMA-based multimodal architecture, the proposed framework substantially improves OCR completeness, multilingual translation accuracy, and robustness under degraded visual conditions. Experimental results on multilingual receipts, menus, posters, signs, handwritten text, and document images demonstrate significantly improved visual-text grounding compared with the baseline model. In particular, the proposed OCR-aware post-training framework improves extraction of small, blurred, spatially scattered, and partially occluded text while reducing reliance on language priors under uncertain OCR conditions. Qualitative comparisons with frontier multimodal systems, including GPT-5-class and Gemini-family models, further suggest improved OCR grounding and reduced hallucination under noisy and visually ambiguous OCR scenarios. Overall, the results indicate that data-centric OCR-aware multimodal post-training provides an effective and scalable direction for improving multilingual OCR and OCR-based visual question answering systems.

Qinwu Xu, Yifan Jiang, Haoyu Ren

Meta AI · UT Austin · Current independent researcher

Multimodal Grounding CoT Reasoning Multimodal Large Language Models Multimodal QA Multimodal Foundation Models Optical Character Recognition+2

May 13, 2026cs.LG

GeoFlowVLM: Geometry-Aware Joint Uncertainty for Frozen Vision-Language Embedding

Standard dual-encoder vision-language models that map images and text to deterministic points on a shared unit hypersphere through

\ell_2

normalization typically expose neither \emph{aleatoric} uncertainty (cross-modal ambiguity) nor \emph{epistemic} uncertainty (lack of training-distribution support). Existing post-hoc methods either recover at most one of the two uncertainty components, or ignore the hyperspherical geometry of these models' embeddings. We propose \textbf{GeoFlowVLM} as a post-hoc adapter that learns the joint distribution of paired

\ell_2

-normalised dual-encoder VLM embeddings on the product hypersphere

\mathbb{S}^{d-1} \times \mathbb{S}^{d-1}

via Riemannian flow matching with a single masked velocity field. A consistency result shows that, in the population limit, the trained network exposes the joint flow and both cross-modal conditional flows as valid Riemannian flow-matching velocity fields on their respective domains. We derive two quantities from this single model: a conditional retrieval entropy that quantifies aleatoric ambiguity with a decision-theoretic interpretation via a Fano-type bound, and a marginal-typicality epistemic score justified by an exact chain-rule decomposition of the joint NLL. This decomposition isolates a cross-modal pointwise-mutual-information term that is structurally discriminative rather than epistemic, and is empirically the only consistently uninformative standalone component. Empirically, the entropy tracks Recall@1 with near-ideal monotonic calibration across three retrieval benchmarks in both directions, and the marginal-typicality sum yields consistently calibrated selective accuracy across four zero-shot classification benchmarks.

Mayank Nautiyal, Li Ju, Andreas Hellander +2

Department of Information Technology, Uppsala University, Uppsala, Sweden · SciLifeLab, Uppsala University, Uppsala, Sweden

Cross-Modal Learning Flow Matching Vision-Language Models Riemannian Flow Matching Vision-Language Retrieval Aleatoric Uncertainty+2

May 13, 2026cs.CV

When Vision Speaks for Sound

Despite rapid progress in video-capable MLLMs, we find that their apparent audio understanding in videos is often vision-driven: models rely on visual cues to infer or hallucinate acoustic information, rather than verifying the audio stream. This issue appears across both state-of-the-art open-source omni models and leading closed-source models from providers such as Google and OpenAI. We characterize this failure mode as an audio-visual Clever Hans effect, in which models appear (falsely) audio-grounded, but actually exploit visual-acoustic correlations without verifying whether the audio and visual streams are truly aligned. To systematically study this behavior, we introduce Thud, an intervention-driven probing framework based on three counterfactual audio edits: Shift, which tests temporal synchronization; Mute, which tests sound existence; and Swap, which tests audio-visual consistency. Beyond diagnosis, we further study a two-stage alignment recipe: intervention-derived preference pairs teach audio verification, while event-level general video preferences regularize the model against over-specialization. Our best 10K-sample recipe improves average performance across the three intervention dimensions by 28 percentage points, while slightly improving performance on general video and audio-visual QA benchmarks.

Xiaofei Wen, Wenjie Jacky Mo, Xingyu Fu +6

dUniversity of California, Davis · pPrinceton University · wUniversity of Wisconsin–Madison +1

Audio-Language Model Evaluation Audio-Visual Understanding Video-Language Model Evaluation Audio Reasoning Audio Understanding Video-Language Models+5

May 13, 2026cs.CL

DiM\textsuperscript{3}: Bridging Multilingual and Multimodal Models via Direction- and Magnitude-Aware Merging

Towards more general and human-like intelligence, large language models should seamlessly integrate both multilingual and multimodal capabilities; however, extending an existing multimodal model to many languages typically requires expensive multilingual multimodal data construction and repeated end-to-end retraining. We study a training-free alternative: injecting multilingual capability into an existing multimodal model by composing residual updates in the shared language model backbone. The key challenge is that multilingual and multimodal updates are heterogeneous, reflecting different functional roles in the shared model. To address this, we propose Direction- and Magnitude-aware Multilingual Multimodal merging (DiM3), which selectively composes the two updates at each parameter dimension while preserving the original vision encoder and multimodal projector. Experiments on multilingual benchmarks in both text-only and vision-language settings, covering 57 languages across LLaVA- and Qwen-based backbones, show that DiM3 consistently outperforms existing merging baselines, substantially improves multilingual performance over the original multimodal model, and remains competitive with dedicated multilingual multimodal fine-tuning while largely retaining general multimodal ability. We further show that DiM3 can be directly applied to already trained multilingual multimodal models and still yield additional gains. Further interpretability analysis shows that DiM3 primarily reshapes intermediate-layer semantic representations, strengthening cross-lingual alignment under both text-only and multimodal inputs while preserving higher-layer task-sensitive structure. Our repository is on https://github.com/wzj1718/DiM3.

Zijing Wang, Mingyang Wang, Ercong Nie +6

Northeastern University, China · CIS, LMU Munich, Germany · Munich Center for Machine Learning (MCML), Germany +1

Multilingual Language Models Multimodal Large Language Models Language Model Merging Multimodal Foundation Models Cross-Lingual Transfer Multimodal Understanding

May 13, 2026cs.LG

SMA: Submodular Modality Aligner For Data Efficient Multimodal Learning

Despite the recent success of Multimodal Foundation Models (FMs), their reliance on massive paired datasets limits their applicability in low-data and rare-scenario settings where aligned data is scarce and expensive. A key bottleneck is the adoption of an instance-level formulation, which learns alignment by maximizing correlation between individual image-text pairs while neglecting the underlying geometric structure across modalities resulting in a modality gap across input modalities. In this paper, we propose a combinatorial paradigm for multimodal alignment that moves beyond pairwise learning and introduce the \emph{Submodular Modality Aligner (SMA)}, which treats multiple augmentations and descriptions of an entity as a set, leveraging multiple descriptions of the data to capture richer cross-modal structure. We instantiate SMA using a principled objective based on Submodular Mutual Information (SMI), which jointly maximizes inter-modality mutual information while reducing cross-modal divergence. This formulation enables the model to effectively utilize multiple positive associations and extract significantly more information from limited data. We evaluate SMA on 14 zero-shot classification and retrieval tasks from the CLIP benchmark and demonstrate consistent gains in the low-data regime. Notably, SMA achieves strong multimodal generalization using only tens of thousands of samples. This is orders of magnitude fewer than standard approaches. Our results highlight the importance of set-based formulations and submodular objectives for data-efficient multimodal learning.

Truong Pham, Anay Majee, Rishabh Iyer

The University of Texas at Dallas

Cross-Modal Learning Cross-Modal Alignment Multimodal Foundation Models Multimodal Contrastive Learning Data-Efficient Learning Multimodal Understanding

May 12, 2026cs.CV

SenseNova-U1: Unifying Multimodal Understanding and Generation with NEO-unify Architecture

Recent large vision-language models (VLMs) remain fundamentally constrained by a persistent dichotomy: understanding and generation are treated as distinct problems, leading to fragmented architectures, cascaded pipelines, and misaligned representation spaces. We argue that this divide is not merely an engineering artifact, but a structural limitation that hinders the emergence of native multimodal intelligence. Hence, we introduce SenseNova-U1, a native unified multimodal paradigm built upon NEO-unify, in which understanding and generation evolve as synergistic views of a single underlying process. We launch two native unified variants, SenseNova-U1-8B-MoT and SenseNova-U1-A3B-MoT, built on dense (8B) and mixture-of-experts (30B-A3B) understanding baselines, respectively. Designed from first principles, they rival top-tier understanding-only VLMs across text understanding, vision-language perception, knowledge reasoning, agentic decision-making, and spatial intelligence. Meanwhile, they deliver strong semantic consistency and visual fidelity, excelling in conventional or knowledge-intensive any-to-image (X2I) synthesis, complex text-rich infographic generation, and interleaved vision-language generation, with or without think patterns. Beyond performance, we show detailed model design, data preprocessing, pre-/post-training, and inference strategies to support community research. Last but not least, preliminary evidence demonstrates that our models extend beyond perception and generation, performing strongly in vision-language-action (VLA) and world model (WM) scenarios. This points toward a broader roadmap where models do not translate between modalities, but think and act across them in a native manner. Multimodal AI is no longer about connecting separate systems, but about building a unified one and trusting the necessary capabilities to emerge from within.

Haiwen Diao, Penghao Wu, Hanming Deng +55

Vision-Language Models Unified Multimodal Models Multimodal Reasoning Multimodal Generation Multimodal Foundation Models Multimodal Understanding

May 12, 2026cs.CV

Fill the GAP: A Granular Alignment Paradigm for Visual Reasoning in Multimodal Large Language Models

Visual latent reasoning lets a multimodal large language model (MLLM) create intermediate visual evidence as continuous tokens, avoiding external tools or image generators. However, existing methods usually follow an output-as-input latent paradigm and yield unstable gains. We identify evidence for a feature-space mismatch that can contribute to this instability: dominant visual-latent models build on pre-norm MLLMs and reuse decoder hidden states as predicted latent inputs, even though these states occupy a substantially different norm regime from the input embeddings the model was trained to consume (Xie et al., 2025; Li et al., 2026; Team et al., 2026). This mismatch can make direct latent feedback unreliable. Motivated by this diagnosis, we propose GAP, a Granular Alignment Paradigm for visual latent modeling. GAP aligns visual latent reasoning at three levels: feature-level alignment maps decoder outputs into input-compatible visual latents through a lightweight PCA-aligned latent head; context-level alignment grounds latent targets with inspectable auxiliary visual supervision; and capacity-guided alignment assigns latent supervision selectively to examples where the base MLLM struggles. On Qwen2.5-VL 7B, the resulting model achieves the best mean aggregate perception and reasoning performance among our supervised variants. Inference-time intervention probing further suggests that generated latents provide task-relevant visual signal beyond merely adding token slots.

Yanting Miao, Yutao Sun, Dexin Wang +8

1Qwen Large Model Application Team, Alibaba · 3Zhejiang University · University of Waterloo +1

Multimodal Large Language Models Latent Visual Reasoning VLM Reasoning Multimodal Foundation Models Multimodal Understanding

May 12, 2026cs.AI

OmniRefine: Alignment-Aware Cooperative Compression for Efficient Omnimodal Large Language Models

Omnimodal large language models (Omni-LLMs) show strong capability in audio-video understanding, but their practical deployment remains limited by high inference cost of long video streams and dense audio sequences. Despite recent progress, existing compression methods for Omni-LLMs typically rely on fixed or native compression units, which can disrupt cross-modal correspondence and the complementary information required for audio-video reasoning, making it difficult to improve inference efficiency while stably preserving performance. To address this, we propose OmniRefine, a training-free two-stage framework for efficient audio-visual token compression in Omni-LLMs. First, Correspondence-Preserving Chunk Refinement refines native chunk boundaries into cross-modally aligned compression units through frame-audio similarity and dynamic programming. Second, Modality-Aware Cooperative Compression jointly compresses video and audio tokens within each refined unit to reduce redundancy while preserving critical evidence. Extensive experiments show that OmniRefine achieves a better efficiency-performance trade-off than strong baselines and maintains stable performance under lower compression ratios. On WorldSense, it still reaches 46.7% accuracy at a 44% token retention ratio, nearly matching the full-token baseline. The code and interface will be released to facilitate further research.

Yuchen Deng, Zidang Cai, Hai-Tao Zheng +3

Tsinghua Shenzhen International Graduate School, Tsinghua University · Pengcheng Laboratory

Audio-Visual Understanding Efficient Multimodal Inference Audio Token Compression Multimodal Large Language Models Multimodal Token Compression Multimodal Foundation Models+2

May 12, 2026cs.MM

Boosting Omni-Modal Language Models: Staged Post-Training with Visually Debiased Evaluation

Omni-modal language models are intended to jointly understand audio, visual inputs, and language, but benchmark gains can be inflated when visual evidence alone is enough to answer a query. We study whether current omni-modal benchmarks separate visual shortcuts from genuine audio-visual-language evidence integration, and how post-training behaves under a visually debiased evaluation setting. We audit nine omni-modal benchmarks with visual-only probing, remove visually solvable queries, and retain full subsets when filtering is undefined or would make comparisons unstable. This yields OmniClean, a cleaned evaluation view with 8,551 retained queries from 16,968 audited queries. On OmniClean, we evaluate OmniBoost, a three-stage post-training recipe based on Qwen2.5-Omni-3B: mixed bi-modal SFT, mixed-modality RLVR, and SFT on self-distilled data. Balanced bi-modal SFT gives limited and uneven gains, RLVR provides the first broad improvement, and self-distillation reshapes the benchmark profile. After SFT on self-distilled data, the 3B model reaches performance comparable to, and in aggregate slightly above, Qwen3-Omni-30B-A3B-Instruct without using a stronger omni-modal teacher. These results show that omni-modal progress is easier to interpret when evaluation controls visual leakage, and that small omni-modal models can benefit from staged post-training with self-distilled omni-query supervision. Project page: https://cheliu-computation.github.io/omni/

Che Liu, Lichao Ma, Xiangyu Tony Zhang +4

StepFun-Audio Team

Cross-Modal Learning Multimodal Large Language Models Multimodal Foundation Models Multimodal Model Evaluation Multimodal Learning Multimodal Understanding

May 12, 2026cs.LG

20/20 Vision Language Models: A Prescription for Better VLMs through Data Curation Alone

Data curation has shifted the quality-compute frontier for language-model and contrastive image-text pretraining, but its role for vision-language models (VLMs) is far less established. We ask how far data curation alone can take VLM performance, holding architecture, training recipe, and compute fixed and varying only the training data. Our pipeline, applied to the MAmmoTH-VL single-image subset, lifts performance by +11.7pp on average across 20 public VLM benchmarks (spanning grounding, VQA, OCR/documents, captioning, spatial/3D, counting, charts, math, brand-ID, and multi-image reasoning) and by +11.3pp on average across all nine capability axes of DatBench, our high-fidelity VLM eval suite. At 2B, our curated model surpasses InternVL3.5-2B by 9.9pp at ~17x less training compute and closes the gap to Qwen3-VL-2B to within 1.8pp at ~87x less compute, from pretraining alone. Beyond accuracy, curation delivers four further properties: (1) Reliability: per-capability std across training seeds drops by ~67% and the lift survives a 4k-to-16k context-length sweep; (2) OOD generalization: the 9-eval OOD average rises by +7.2pp, and multi-image BLINK rises by +3.09pp despite single-image-only training, with Visual Correspondence gaining +11.8pp; (3) Behavioral gains beyond benchmarks: across ~1,100 open-ended queries the curated 2B is more honest and more specific than the matched-compute baseline, and more concise and less refusal-prone than a frontier 2B reference; (4) Pareto-dominance on inference cost: at every scale (1B, 2B, 4B) the curated model raises accuracy while lowering response FLOPs vs. the matched-compute baseline, and the curated 4B matches near-frontier accuracy at 3.3x lower response FLOPs than Qwen3-VL-4B. Data curation is a high-leverage tool for building better VLMs, reaching near-frontier accuracy at up to ~150x less training compute.

DatologyAI, :, Siddharth Joshi +32

DatologyAI Team1

Training Data Curation Large Vision-Language Models Multimodal Foundation Models Vision-Language Pretraining

May 11, 2026cs.AI

LatentRouter: Can We Choose the Right Multimodal Model Before Seeing Its Answer?

Multimodal large language models (MLLMs) have heterogeneous strengths across OCR, chart understanding, spatial reasoning, visual question answering, cost, and latency. Effective MLLM routing therefore requires more than estimating query difficulty: a router must match the multimodal requirements of the current image-question input with the capabilities of each candidate model. We propose LatentRouter, a router that formulates MLLM routing as counterfactual multimodal utility prediction. Given an image-question query, LatentRouter extracts learned multimodal routing capsules, represents each candidate MLLM with a model capability token, and performs latent communication between these states to estimate how each model would perform if selected. A distributional outcome head predicts model-specific counterfactual quality, while a bounded capsule correction refines close decisions without allowing residual signals to dominate the prediction. The resulting utility-based policy supports performance-oriented and performance-cost routing, and handles changing candidate pools through shared per-model scoring with availability masking. Experiments on MMR-Bench and VL-RouterBench show that LatentRouter outperforms fixed-model, feature-level, and learned-router baselines. Additional analyses show that the gains are strongest on multimodal task groups where model choice depends on visual, layout-sensitive, or reasoning-oriented requirements, and that latent communication is the main contributor to the improvement. The code is available at: https://github.com/LabRAI/LatentRouter.

Xueqi Cheng, Yushun Dong

Department of Computer Science Florida State University University

Efficient Multimodal Inference Multimodal Large Language Models LLM Routing Multimodal Foundation Models Multimodal Understanding

May 11, 2026cs.CV

HiDream-O1-Image: A Natively Unified Image Generative Foundation Model with Pixel-level Unified Transformer

The evolution of visual generative models has long been constrained by fragmented architectures relying on disjoint text encoders and external VAEs. In this report, we present HiDream-O1-Image, a natively unified generative foundation model via pixel-space Diffusion Transformer, that pioneers a paradigm shift from modular architectures to an end-to-end in-context visual generation engine. By mapping raw image pixels, text tokens, and task-specific conditions into a single shared token space, HiDream-O1-Image achieves a structural unification of multimodal inputs within an Unified Transformer (UiT) architecture. This native encoding paradigm eliminates the need for separate VAEs or disjoint pre-trained text encoders, allowing the model to treat diverse generation and editing tasks as a consistent in-context reasoning process. Extensive experiments show that HiDream-O1-Image excels across various generation tasks, including text-to-image generation, instruction-based editing, and subject-driven personalization. Notably, with only 8B parameters, HiDream-O1-Image (8B) achieves performance parity with or even surpasses established state-of-the-art models with significantly larger parameters (e.g., 27B Qwen-Image). Crucially, to validate the immense scalability of this paradigm, we successfully scale the architecture up to over 200B parameters. Experimental results demonstrate that this massive-scale version HiDream-O1-Image-Pro (200B+) unlocks unprecedented generative capabilities and superior performance, establishing new state-of-the-art benchmarks. Ultimately, HiDream-O1-Image highlights the immense potential of natively unified architectures and charts a highly scalable path toward next-generation multimodal AI.

Qi Cai, Jingwen Chen, Chengmin Gao +22

Diffusion Transformer Text-Guided Image Editing Unified Multimodal Models Pixel-Space Diffusion Models Multimodal Foundation Models T2I Diffusion Models+1

May 11, 2026cs.CV

Qwen-Image-2.0 Technical Report

We present Qwen-Image-2.0, an omni-capable image generation foundation model that unifies high-fidelity generation and precise image editing within a single framework. Despite recent progress, existing models still struggle with ultra-long text rendering, multilingual typography, high-resolution photorealism, robust instruction following, and efficient deployment, especially in text-rich and compositionally complex scenarios. Qwen-Image-2.0 addresses these challenges by coupling Qwen3-VL as the condition encoder with a Multimodal Diffusion Transformer for joint condition-target modeling, supported by large-scale data curation and a customized multi-stage training pipeline. This enables strong multimodal understanding while preserving flexible generation and editing capabilities. The model supports instructions of up to 1K tokens for generating text-rich content such as slides, posters, infographics, and comics, while significantly improving multilingual text fidelity and typography. It also enhances photorealistic generation with richer details, more realistic textures, and coherent lighting, and follows complex prompts more reliably across diverse styles. Extensive human evaluations show that Qwen-Image-2.0 substantially outperforms previous Qwen-Image models in both generation and editing, marking a step toward more general, reliable, and practical image generation foundation models.

Bing Zhao, Chenfei Wu, Deqing Li +72

T2I Generation Compositional T2I Generation Multimodal Foundation Models Diffusion-Based Image Editing Multimodal Diffusion Models

May 11, 2026cs.CL

Phoenix-VL 1.5 Medium Technical Report

We introduce Phoenix-VL 1.5 Medium, a 123B-parameter natively multimodal and multilingual foundation model, adapted to regional languages and the Singapore context. Developed as a sovereign AI asset, it demonstrates that deep domain adaptation can be achieved with minimal degradation to broad-spectrum intelligence and alignment. Continued pretraining was performed on Mistral Medium 3.1 using a localized 1-trillion tokens multimodal corpus, followed by a 250-billion tokens long-context extension phase. Subsequent post-training incorporated a novel human-annotated Singapore multimodal dataset and curated textual corpus on Singapore culture, knowledge, and legislation, totaling 22-billion tokens. An additional 5 billion tokens of model alignment was performed through Online Direct Preference Optimization. Phoenix-VL 1.5 Medium achieves state-of-the-art performance for its size on Singapore multimodal, legal, and government policy benchmarks while remaining globally competitive on general multimodal intelligence, multilingual, and STEM benchmarks. We also introduce a novel evaluation suite encompassing localized knowledge benchmarks and an institutionally aligned model behavior and safety framework. We report the data curation principles, training methodology, and highlight benchmark and inference performance.

Team Phoenix, :, Arka Ray +30

HTX × Mistral AI ∗

Multilingual Language Models Domain Adaptation Multimodal Pretraining Multimodal Large Language Models Domain Adaptation for LLMs Multimodal Foundation Models+1

May 11, 2026cs.CV

The Cartesian Shortcut: Re-evaluate Vision Reasoning in Polar Coordinate Space

As current Multimodal Large Language Models rapidly saturate canonical visual reasoning benchmarks, a key question emerges: do these strong scores genuinely reflect robust visual understanding? We identify a pervasive vulnerability, the Cartesian Shortcut: models frequently discretize the orthogonal grid-based layouts prevalent in visual reasoning benchmarks into explicit textual coordinates, offloading reasoning from visual perception to text-based deduction. To re-evaluate visual reasoning when this shortcut is unavailable, we introduce Polaris-Bench, which re-formulates 53 visual reasoning tasks in Polar coordinate space with paired Cartesian counterparts that preserve task semantics, disrupting the orthogonal structure that models exploit. Comprehensive evaluation across

14

state-of-the-art MLLMs reveals that frontier models achieving

69

--

83\%

on Cartesian layouts collapse to

31

--

39\%

on Polar equivalents. Moreover, thinking gains largely vanish on Polar layouts, prompting interventions fail to close the gap, and comparable drops arise on other non-orthogonal layouts. These findings reveal that current MLLMs' visual reasoning performance is strongly coupled to orthogonal grid structure, a fragility that is consistent across model families but far smaller in humans, who maintain 88.8% accuracy on Polar layouts. The benchmark, evaluation suite, and leaderboard are publicly available at https://google-deepmind.github.io/polaris-bench.

Xia Hu, Zhenrui Yue, Brian Potetz +4

Google DeepMind · Stanford University · Google Research

VLM Evaluation Spatial Reasoning Benchmarks Vision-Language Models Visual Reasoning Visually Grounded Reasoning Multimodal Foundation Models+1