The paper introduces Align Then Reason (ATR), a multilingual lip‑sync judge that aligns frame‑level lip representations with phonetic units of a candidate text line and then uses a language model to evaluate both content and timing. ATR achieves significant improvements over existing baselines on a seven‑language benchmark, with mean AUC gains of up to 59.4% for 2B reasoners and similar gains across other LLM families. The method also transfers well to unseen languages and outperforms lip‑reading baselines on real dubbing tasks such as dub‑line reranking and script‑to‑clip assignment.
By Rui Liu, Bhavin Jawade, Haoqi Li, Shivam Mehta, Karan Saxena, Yinghong Lan, Cameron R. Wolfe
Watch Your Speech (WYS) is a video‑to‑speech synthesis framework that uses textual conditioning to resolve the one‑to‑many mapping problem inherent in silent talking‑face videos. It fuses textual context with video sequences via an attention‑based embedding module and employs a conditional flow matching objective to produce high‑fidelity, phonemically accurate speech. Experiments on LRS2 and LRS3 show that WYS sets new state‑of‑the‑art results in audio‑visual synchronization while maintaining competitive word‑error rates, and subjective tests confirm near‑human naturalness.
By Gunwoo Lee, Yoori Oh, Yoseob Han
The paper presents a framework for creating a scalable synthetic dataset of controllable video interactions by generating explicit start and end state images using image editing models. It introduces State‑Guided Sampling (SGS) to produce seamless videos anchored on these states, reducing artifacts seen in naive conditional generation. An automated evaluation system aligned with human judgments is also developed, and experiments demonstrate that fine‑tuning a base model on this dataset markedly improves its ability to generate plausible interactions.
By Jiho Jang, Jinyoung Kim, Nojun Kwak, Kyungjune Kim
Ego2Act is a new benchmark that tests video generation models on goal‑directed, egocentric manipulation tasks. It contains 2,640 videos from 110 real‑world tasks, each requiring multiple steps of object manipulation to achieve a high‑level goal. The benchmark includes Ego2ActJudge, a reference‑free evaluation pipeline that better aligns with human judgments of task completion and physics plausibility.
By Patrick Amadeus Irawan, Iskandar Muda Rizky Parlambang, Rava Maulana, Qinrong Cui, Erland Hilman Fuadi, Zayd M. K. Zuhri, Nanda Ryaas Absar, Ahmed Elshabrawy, Wilfried Ariel Mulyawan, Shoubin Yu, Yue Zhang, Mohit Bansal, Alham Fikri Aji
PhysicsLENS is a new dataset and benchmark designed to evaluate how well video generation models capture physical properties relevant to robotics. It consists of matched scenario pairs that share the same conditioning frame and task but differ in underlying physics, and it covers seven physical domains such as collision, gravity, and friction. The benchmark includes over 400 human-annotated labels from four video generation models, revealing that many plausible-looking videos still ignore the specified physical property.
By Isaiah Milkey, Som Sagar, Aditya Taparia, Xinyuan Liu, Jiqing Wen, Ransalu Senanayake
Dyna3 is a training‑free framework that extends the depth foundation model DA3 to perform 4D dynamic scene reconstruction without fine‑tuning. By leveraging DA3’s cross‑view features and a best‑match search, it distinguishes static surfaces from moving objects, and uses vision‑language models to generate semantic prompts for SAM 3 to achieve precise instance‑level segmentation. Experiments on four datasets show Dyna3 outperforms correspondence‑trained methods, improving dynamic object segmentation by +5.5 pp, speeding pose estimation 13×, and reducing memory usage 4–8×.
By Xinhao Xiang, Weiyang Li, Zhijie Zheng, Abhijeet Rastogi, Jiawei Zhang
FiVOS is an interactive video object segmentation algorithm tailored for fish segmentation in aquaculture. It introduces a mask block filter and serial noise filters to detect and correct erroneous propagated masks early, improving accuracy and robustness. Experiments show FiVOS achieves state‑of‑the‑art performance on two newly constructed fish‑specific datasets.
By Yuqing Duan, Song Zhang, Shili Zhao, Daoliang Li, Ran Zhao
FedCKA introduces a Centered Kernel Alignment (CKA)-based method for federated 3D perception that dynamically balances personalization and globalization. By computing layer-wise feature similarities between local client models and a global consensus model, FedCKA generates client‑specific aggregation masks to selectively share representation‑consistent layers. Experiments on a unified multi‑domain nuScenes benchmark demonstrate that FedCKA surpasses established federated baselines, improving average NDS by 7 percentage points.
By Jolle Verhoog, Ali Burak \"Unal, Holger Caesar
PAGER is a label‑free adaptation method that aligns partial, viewpoint‑dependent 3D observations with a frozen global semantic space. It uses matched‑point feature alignment and relational supervision to anchor partial features to their global counterparts while preserving similarity structure, all without altering the pretrained encoder or global probe. Experiments show PAGER outperforms label‑supervised PEFT on Sonata and Concerto, and achieves superior zero‑shot transfer from ScanNet to ScanNet++ compared to fully fine‑tuned Sonata.
By Akira-Miranda Adeyomi Adeniran-Lowe, Binod Singh, Lars Arnold Dethlefsen, Lazaros Nalpantidis, Theodora Kontogianni
The paper introduces a Multi-layer Fusing Transformer that uses cross‑attention to merge visual and textual features across multiple layers, enabling extraction of information from low to high levels. Experiments and ablation studies on the ViVQA dataset demonstrate that this architecture outperforms competitive baselines for Vietnamese Visual Question Answering. The work addresses a notable gap in VQA research for Vietnamese, a language with limited prior resources.
By Cong Phu Nguyen, Huy Tien Nguyen, Tung Le
Unified models trained for both instruction-based image editing and text-to-image generation typically keep source-image conditioning throughout denoising. This study shows that for some edits, source attention naturally decreases during sampling, suggesting that switching to text-to-image conditioning for short intervals can improve edit quality without sacrificing perceptual preservation. Across three editors and four benchmarks, task switching balances quality and preservation, demonstrating that unified editors benefit from using both conditioning modes they are trained for.
By Lidia Troeshestova, Alexander Ustyuzhanin, Sergey Kastryulin
GenCOPE introduces a synthetic-to-real (Syn2Real) approach for category-level object pose estimation (COPE) that eliminates the need for labor-intensive real-world data collection. By learning domain-invariant representations through 2D and 3D semantic consistency constraints and employing an end-to-end pose regression framework with 2D-3D cross consistency, the model achieves robust generalization across synthetic and real domains. The architecture relies solely on global features, resulting in a lightweight and efficient design validated on REAL275, Wild6D, and real-world robotic manipulation scenes.
By Jian Liu, Wei Sun, Zhenqi Dai, Hui Yang, Jian Xiao, Nicu Sebe, Na Zhao
MapLightning introduces a new approach to online vectorized HD map construction that replaces dense bird’s‑eye‑view grids with a compact set of 1D learnable map tokens. Using a transformer‑based mapper that applies full self‑attention to combined image and map tokens, the method achieves efficient inference, lower memory usage, and improved global context without relying on camera projection parameters. Experiments on nuScenes and Argoverse 2 show state‑of‑the‑art accuracy, with a lightweight variant outperforming MapTRv2 by over 10 mAP and running 1.73× faster at 40+ FPS while using 53 % less memory.
By Shen Zheng, Anurag Ghosh, Mani Ramanagopal, Srinivasa Narasimhan
DiDE is a training‑free framework that enables independent control of color and texture when stylizing 3D assets generated by rectified flow‑based image‑to‑3D models. It leverages the overcomplete latent space of these models, partitioning channels so that texture occupies a small subset while color is encoded in a free subspace. Experiments on the newly collected Disen3D‑Bench benchmark demonstrate that DiDE outperforms existing 2D and 3D stylization methods in color fidelity, texture transfer, and content preservation.
By Tao Wu, Alexandra Gomez-Villa, Senmao Li, Yaxing Wang, Joost van de Weijer, Kai Wang
The paper introduces a semi‑supervised student–teacher framework for restoring images degraded by adverse weather. It uses an unreliable database of failed teacher predictions as negative samples for contrastive learning and a reliable database of high‑quality predictions as positives, jointly improving restoration quality. A phase‑spectrum semantic constraint and an adaptive phase consistency loss replace expensive text supervision and balance guidance based on degradation severity, leading to superior performance on real‑world benchmarks.
By Cap Dang Xuan Kiet, Tat-Jen Cham
World Observer introduces a joint actor‑observer framework for persistent world modeling, decoupling the agent’s view from panoramic observers that track regions outside the actor’s field of view. By warping from a shared panoramic source and using an Observer Sink for high‑resolution references, the method maintains accurate object states and dynamics even when they leave and re‑enter the actor’s view. The approach is evaluated with new world‑space metrics and a benchmark across real and synthetic scenes, showing improved out‑of‑view dynamics while preserving visual fidelity, camera control, and 3D adherence.
By Hyunwook Choi, Dahyun Chung, Hyunsung Kim, Siyoon Jin, Jinhyeok Choi, Junyoung Seo, Seungryong Kim
UnifiedAttack introduces a benchmark for testing the safety of large multimodal models (LMMs) in tasks that combine text and image to produce harmful content. The benchmark focuses on the additional harm that arises from cross‑modal synergy and includes filtered multimodal samples and synthesized disinformation queries. A synergistic hijacking framework—comprising In‑Context Reskinning (ICR) and Cognitive Planning Injection (CPI)—is proposed to expose vulnerabilities, and extensive evaluations show that UnifiedAttack consistently bypasses current alignment defenses in state‑of‑the‑art architectures.
By Bingjun Luo, Jialin Guo, Tony Wang, Siqi Li
The paper introduces Disentangled Diffusion Autoencoders (DiDAE), a method that wraps a frozen foundation model with a conditional diffusion decoder to generate counterfactual edits along disentangled dictionary directions. DiDAE can use supervised or unsupervised dictionaries, requires no gradients, and is up to 2000 times faster than existing approaches. Evaluations on six datasets show that its counterfactuals match or surpass state‑of‑the‑art methods and can repair downstream classifiers via Counterfactual Knowledge Distillation, with the entire workflow available as an open‑source library.
By Sidney Bender, Benedikt Kunz, Ahmed Zeid, Shinichi Nakajima, Klaus-Robert M\"uller, Marco Morik
The paper investigates why post‑hoc out‑of‑distribution (OOD) detectors often report inflated gains. It shows that many detectors learn to recognize the specific dataset they were trained on rather than true novelty, a phenomenon called the inflation gap. By holding out entire OOD datasets during evaluation, the authors demonstrate that most reported gains vanish, and they provide a closed‑form method to predict which fits will inflate.
By Donghoon Lee, Shinjin Kang
The paper introduces VITA, a multi‑source vicinal transfer augmentation method designed to improve out‑of‑distribution generalization in computer vision. VITA combines tangent transfer, which generates initial augmented samples to enhance robustness against various image corruptions, with an integration step that uses a generative model to produce on‑manifold samples from multiple vicinal sources. Experiments on corruption benchmarks show that VITA outperforms existing state‑of‑the‑art augmentation techniques.
By Minghui Chen, Cheng Wen, Feng Zheng, Fengxiang He, Ling Shao