Chart2SVG is a multimodal large language model that transforms static raster chart images into editable SVGs enriched with semantic structure. By embedding chart‑specific semantic tokens into a vision‑language framework and training on the Beagle+ dataset of 33K distilled chart samples, the model captures both geometric primitives and their functional roles. The resulting SVGs are visually accurate and structurally consistent, and the accompanying Chart Structure Graph (CSG) exposes visual dependencies for interactive exploration, chart repurposing, and layout reuse.
By Jinning Cui, Lu Chen, Haoyan Shi, Yue He, Chenglong Wang, Mengyu Zhou, Weidong Huang, Yunhai Wang
The paper presents a method for accurately measuring 3D and 2D circular centers, a key geometric sensing problem for instruments like cameras and LiDARs. It integrates a conformal‑geometric‑algebra estimator with RANSAC to jointly recover the 3D center, normal, and radius from noisy LiDAR points, and uses a chord‑length‑variance criterion to estimate the 2D projected center, resolving ambiguity with homography validation or a quasi‑RANSAC fallback. Experiments on synthetic and real sensors demonstrate improved center measurement accuracy and reduced extrinsic calibration error across various target and sensor setups.
By Jiajun Jiang, Xiao Hu, Wancheng Liu, Wei Jiang
The paper demonstrates that vision‑language models (VLMs) possess a small set of attention heads, called Visual Retrieval Heads (VRHs), that are causally responsible for linking text prompts to specific image regions. By adapting head‑scoring techniques from language models, the authors identify VRHs as the heads whose attention from output prediction tokens, summed over the ground‑truth referent region, most reliably indicates causal grounding. Experiments across eleven VLMs and five referring‑expression benchmarks show that masking the top 20 VRHs can drop grounding accuracy by up to 80 percentage points, while random masking has little effect, and that VRHs generalize across diverse visual tasks and transfer across models sharing an LLM backbone.
The paper introduces Reader-Centric Misleadingness Understanding (RCMN), a framework that defines misleadingness in public discourse through five dimensions: misleading mechanism, likely reader interpretation, evidence-warranted interpretation, emotional arousal, and communicative intent. Using this framework, the authors build an evidence-grounded dataset and find that misleadingness is diverse, extending beyond fabrication to include unsupported inference, exaggeration, and omission, often linked to heightened emotional arousal and distortive intent. They also evaluate whether lightweight claim-and-context representations can capture reader-level interpretations, showing that while such representations can recover interpretations, they struggle to identify the underlying misleading mechanisms, underscoring the need for richer contextual and evidential data.
The safety of large vision-language models is increasingly stress-tested by multimodal jailbreaks, yet existing attacks remain largely static at the meta level: template-based attacks freeze the image...
Internet memes are a pervasive form of multimodal online communication; however, such communication often involves users from diverse linguistic and cultural backgrounds. Therefore, adapting memes acr...
The paper explores how AI can develop its own aesthetic categorization of art across text, audio, image, and video without explicit labels. Using a self‑supervised framework, the authors embed these modalities into a shared 256‑dimensional space and iteratively cluster the data to uncover aesthetic structure. They compare the AI’s cluster assignments with human affective labels, highlighting divergences and discussing implications for cross‑modal similarity, media organization, and automated labeling.
MVC-Bench is a calibration-focused benchmark for medical vision‑language models, evaluating how well these models express confidence across different modalities, backbones, and domain shifts. It tests robustness to modality, backbone, and domain changes, the effectiveness of calibration and prompt‑tuning strategies, and stability under prompt‑template and random‑seed variations. The benchmark includes 1638 experiments, reporting accuracy and Expected Calibration Error (ECE) along with other calibration metrics, and introduces a simple train‑time calibration method, Multi‑Class Margin (MCM) regularization, that achieves the lowest ECE in most settings.
GraphMemix introduces a combinatorial‑optimization graph memory framework that organizes long‑term multimodal agent memory as query‑aware evidence forests. It constructs candidate graphs by expanding seed memories through schema and semantic relations, then decouples evidence utility from anchor‑conditioned relation verification to reduce redundancy, and finally optimizes a forest‑format memory context within a maximum evidence budget. Experiments on four benchmarks show significant accuracy gains and a new Pareto frontier between accuracy and lifecycle cost.
4DSynth is a controllable procedural system that transforms natural-language descriptions, blueprint masks, or single photographs into editable 4D environments featuring explicit geometry, animated actors, collision-free trajectories, and physics-ready simulation states. The system unifies animation, camera planning, rendering, and task generation within a single geometry-grounded representation, enabling scalable creation of diverse, interactive scenes. Using 4DSynth, the authors built 4DSynth-Nav, an interactive navigation benchmark that demonstrates the reproducibility of failures and tunable difficulty across three tiers for vision‑language models.
The paper presents a method for linking written English keywords to spoken Hindi utterances using only visual grounding. By extracting English tags from images with captioning systems and aligning Hindi speech segments via self‑supervised features, the authors build an alignment‑based model that outperforms previous attention‑based neural approaches. The study also shows that adding negative examples improves alignment quality, demonstrating that cross‑lingual word‑to‑speech mappings can be learned without transcriptions or explicit training.
SAGE is an end‑to‑end CLIP‑based framework that augments vision‑language time‑series forecasting by jointly modeling temporal, cross‑variable, textual, and visual information. It processes frequency‑enhanced patches and variable tokens through a CLIP text encoder, while gated residual paths inject variable‑specific descriptions and statistical descriptors. A frozen CLIP vision encoder aligns rendered series with temporal representations via a training‑only contrastive objective, enabling multimodal alignment and variable‑level knowledge without using an LLM during inference.
The paper introduces the Multi-Instruction Multi-Shot Long-Video Editing (MMLVE) task, aiming to edit long videos with multiple instructions while maintaining consistency across shots, decoupling instructions, and preserving spatiotemporal structure. It proposes an agentic editing framework that combines Large Language Models and Vision‑Language Models for shot-level decoupling and precise instruction parsing. A new dataset, MMLVE‑Bench, and three evaluation metrics are created to benchmark this task, and experiments show the proposed MMLVE‑Agent outperforms existing closed‑source state‑of‑the‑art methods by eliminating hallucinations and ensuring seamless transitions.
The paper introduces Instruct-to-Act, a system that decouples high‑level planning from low‑latency control by combining a vision‑language model (VLM) planner with a world‑model controller. The VLM generates sparse, high‑level text instructions, while the controller executes them autonomously at high frequency. Experiments across seven embodied environments, including multi‑agent settings, show that this approach outperforms both controller‑only and direct VLM action‑generation methods, maintains fast control, and allows swapping in different pretrained VLM planners without fine‑tuning.
G2D is a training‑free framework that combines a discriminative model (CLIP) for broad candidate retrieval with a generative vision‑language model for fine‑grained, image‑grounded verification. By using CLIP’s top‑K shortlist and a structured prior from candidate names and probabilities, G2D focuses generative reasoning on uncertain samples, employing fixed confidence routing, entropy‑adaptive candidate sizing, and trie‑constrained decoding to produce a single valid output. Across eight benchmarks, G2D achieves an average accuracy of 68.85%, outperforming both CLIP (59.35%) and the standalone generative model (63.11%), and it also transfers effectively to other models such as DCLIP, WaffleCLIP, and CuPL.
The paper introduces a new task called compositional layout understanding, focusing on interpreting complex, multi-layer document and UI layouts that involve hierarchical relationships among visually entangled elements. It presents CoDeLayout, a VQA dataset of about 20,000 real-world layouts annotated with compositional element pairs and design intent, and identifies two main challenges for current vision‑language models: semantic drift between textual metadata and visual content, and structural ambiguity in hierarchical inter‑element relationships. To address these, the authors propose MASON, a post‑training paradigm that combines multimodal alignment and structural perception, achieving a 91.66% accuracy on CoDeLayout and outperforming full‑data direct fine‑tuning with only 30% of the training data.
The paper introduces DPA-I2P, a depth‑guided projective alignment method for image‑to‑point‑cloud registration in autonomous driving. It employs Ray‑Conditioned Metric Depth Encoding and Projection‑Consistent Vision Lifting to align depth and visual cues geometrically, and uses Cross‑Modal Query Pruning to filter unreliable matches during refinement. Experiments on KITTI and nuScenes show significant improvements, reducing rotation and translation errors by up to 55.6% compared to existing implicit baselines.
VBVR-Pro is a closed‑loop testbed that enables native visual reasoning through generation, offering 300 procedurally generated tasks that scale training and allow strong transfer to external benchmarks. It supplies verifiable reward scorers based on deterministic, task‑specific rules, outperforming VLM‑as‑a‑judge approaches and providing reliable signals for reinforcement learning. The suite also facilitates controlled modality studies, revealing that video generation excels at persistent spatiotemporal tracking while interleaved generation offers a compute‑efficient alternative, and highlights the importance of vision‑native trajectories for reasoning.
By Junxiang Xu, Ruisi Wang, Fanyi Pu, Maijunxian Wang, Ran Ji, Tongxi Zhou, Chenyang Gu, Jing Zuo, Hongcan Xiao, Yimeng Geng, Wanqi Yin, Wei Chen, Oscar Qian, Zhengan Yan, Ziqi Huang, Haiwen Diao, Liang Pan, Bo Li, Xiangyu Fan, Dezhi Luo, Fengyuan Yu, Zehong Zhao, Qingying Gao, Tinghui Zhu, Yilan Zhang, Jingqi Tong, Pinyuan Feng, Zhengze Jiang, Letian Wang, Ziyu Guo, Renrui Zhang, Jieneng Chen, Sonia Joseph, Constantin Venhoff, Saman Motamed, Mengyue Yang, Chandra Sripada, Alan Yuille, Philip Torr, Lvmin Zhang, Vikash Kumar, Daniel Khashabi, Nikolaus Kriegeskorte, Rapha\"el Milli\`ere, Vincent C. M\"uller, Anyi Rao, Quan Wang, Ziwei Liu, Dahua Lin, Lei Yang, Hokin Deng, Zhongang Cai
The paper introduces Clue-OPSD, a clue‑privileged on‑policy self‑distillation framework that improves long‑video understanding by focusing on short, question‑relevant clue intervals rather than the entire video. Experiments on multiple benchmarks and Qwen3.5 model scales show that this approach consistently outperforms standard backbone models and competes strongly with supervised post‑training baselines, all while requiring fewer input frames and no additional inference modules.
By Kaishen Wang, Dongdi Zhao, Yijun Liang, Dingqiang Ye, Ruibo Chen, Heng Huang, Di Fu