The paper presents a zero‑shot framework for high‑throughput generative video super‑resolution (VSR) that operates on a single high‑end consumer GPU without retraining. It decouples spatial upscaling, handled by a time‑distilled one‑step diffusion backbone, from temporal stabilization, achieved with a lightweight pretrained flow‑based interpolator. A resolution‑aware hybrid attention scheme replaces high‑resolution global attention with neighborhood attention to improve scalability, achieving up to ~32 FPS at 720p and a 44% speedup at 4K on an NVIDIA RTX 4090, while maintaining competitive perceptual quality and temporal consistency on standard VSR benchmarks.
By Georgios Panaretou, Sotirios Chatzis
MoE-ViE introduces a Mixture-of-Experts vision encoder that scales efficiently for image and video understanding, outperforming dense counterparts across various sizes. The study shows fine‑grained MoE topologies provide significant gains, and proposes an auxiliary‑loss‑free balancing variant and a specialized MoE kernel to reduce inference latency. With frame‑level distillation and a novel freezing mechanism, the largest MoE‑ViE model matches state‑of‑the‑art zero‑shot performance while being 1.7× larger and 76% faster, and it outperforms other encoders when paired with a language model on both image and video benchmarks.
VETO (Video Efficient Token Optimization for Vision Language Models) is a plug‑in that reduces the quadratic cost of visual tokens in long‑video inference by applying dual‑axis compression: an intra‑frame compressor merges semantically similar tokens within each frame, and an inter‑frame compressor merges temporally redundant frames. By first compressing spatial dimensions, VETO lowers the cost of subsequent global temporal matching, surpassing single‑axis methods and achieving up to 45% faster inference on models such as LLaVA‑OneVision‑7B while maintaining or improving accuracy. The approach is universally applicable across LLaVA‑OneVision, InternVL‑2.5, and LongVA, preserving or enhancing zero‑shot accuracy even under extreme token budgets.
By Gueter Josmy Faure, Hao Ping Wang, Min-Hung Chen, Winston H. Hsu
arXiv:2607. 00687v1 Announce Type: cross Abstract: Comparing transformer backbones for image segmentation is confounded: each is paired with a different decoder, recipe, and pretraining, so reported differences rarely reflect the backbone itself.
By Tobias Christian Nauen, Anosh Billimoria, Federico Raue, Stanislav Frolov, Brian B. Moser, Andreas Dengel
arXiv:2607. 24570v1 Announce Type: cross Abstract: Large-scale video platforms process millions of uploads hourly, requiring moderation systems that can localize when and where policy violations occur within each video.
By Jiameng Zhang, Srikanth Madikeri
arXiv:2605.08158v2 Announce Type: replace-cross
Abstract: Long video understanding with multimodal language models suffers from three compounding bottlenecks: heavy decode cost to obtain dense RGB fr...
By Haopeng Jin, Wenlong Zhao, Tao Yu, Tiankun Yang, Zhenyu Guan, ShiQuan Dong, Yujia Yang, Shani Ye, Chubin Chen, Hongzhu Yi, Jinwen Luo
arXiv:2606.06158v2 Announce Type: replace
Abstract: Adaptive video tokenisation seeks to dynamically allocate token budgets based on the underlying visual complexity of a sequence. Current continuous...
By Kevin Dave, Sai Aditya Patkuri, Chhaya Kumar Das, Gouranga Bala, Rajeshkumar SA, R. Venkatesh Babu
The paper introduces Parallel Tube Decoding (PTD), a generative approach for spatio‑temporal video grounding that splits the task into a temporal block and simultaneous time‑conditioned spatial blocks, eliminating token‑level and trajectory‑level dependencies. PTD uses Decoupled Block Attention to allow parallel spatial generation while maintaining shared video‑query context, and incorporates localization‑aware policy optimization for temporal boundaries and spatial geometry. Experiments on VidSTG show PTD cuts tube completion latency by 79× and boosts spatial decoding throughput by 92× compared to autoregressive decoding, while improving grounding accuracy and performing well on related tasks such as temporal grounding, VideoQA, and referring video object tracking.
By Hanoona Rasheed, Haania Siddiqui, Ming-Hsuan Yang, Fahad Shahbaz Khan, Salman Khan
arXiv:2510. 09608v2 Announce Type: replace-cross Abstract: Vision-language models (VLMs) could power real-time assistants and autonomous agents, but they face a critical challenge: understanding near-infinite video streams without escalating latency and memory usage.
By Ruyi Xu, Guangxuan Xiao, Yukang Chen, Liuning He, Yao Lu, Song Han
arXiv:2610.20137v1 Announce Type: new
Abstract: Current video object removal methods suffer from prohibitive latency (minutes per video) and non-causal processing that prevents real-time interaction....
By Zhuobai Dong, Xiaobing Tu, Xiantao Zhang, Jinkui Ren, Yinggui Wang, Yue Ma, Linfeng Zhang
The paper introduces Event-Driven Refresh + Recurrence Memory (EDRRM) to improve Referring Video Object Segmentation (RVOS). EDRRM selectively re-invokes the Sa2VA model at stable change points, using an EMA‑smoothed event score from tracking cues and a recurrence memory that retrieves anchor frames via CLIP similarity. Experiments on Ref‑DAVIS17, MeViS, and ReVOS show that EDRRM maintains or surpasses J&F scores while reducing refresh calls and false‑positive failures, with modest overhead compared to Sa2VA inference.
By Abu Hanif Muhammad Syarubany, Jaehyun Jang, Siwoo Lim, Seungyeon Ryu, Chang D. Yoo
arXiv:2609.37001v1 Announce Type: cross
Abstract: Diffusion Transformers (DiTs) enable high-quality video generation but suffer from substantial inference latency, primarily attributable to the compu...
By Xingyu Jia, Baole Ai, Ang Wang, Kang Zhao, Yong Li