OmniVBench introduces a comprehensive benchmark and a large-scale dataset for omni reference-to-video (R2V) generation, addressing gaps in existing evaluations that focus only on limited reference types and holistic consistency. The benchmark expands evaluation across 7 task families and 18 fine-grained tasks, covering content, motion, style, structure, narrative, and multi-reference settings, and employs a factor‑grounded evaluation with 12,172 checklist items to assess preservation, disentanglement, and routing of reference factors. The accompanying Omni‑R2V Dataset provides 340K training samples derived from professional video footage, along with task‑specific pipelines for scalable data construction, enabling broader research and revealing performance gaps in current R2V models.
arXiv:2604. 06010v2 Announce Type: replace Abstract: Video fundamentally intertwines two crucial axes: the dynamic content of a scene and the camera motion through which it is observed.
RefVideo-6M is a new large-scale reference-guided editing dataset that includes 5 million video editing samples and 1 million image editing samples, each paired with about 6 million visual references. The dataset is constructed to avoid artifacts by using real, artifact‑free videos as targets and filtering input conditions with multiple editing experts, thereby providing reliable supervision. It enables models to learn fine‑grained visual correspondence beyond text‑only instructions and supports the training of a reference‑guided video editing model, Ref‑MoT, which shows improved visual quality, controllability, and reference consistency.
arXiv:2608. 05745v1 Announce Type: cross Abstract: Video Virtual Try-On (VVT) synthesizes a video of a person wearing a target garment while preserving identity, motion, and scene dynamics.
arXiv:2501.04001v4 Announce Type: replace Abstract: This work presents Sa2VA, the first comprehensive, unified model for dense grounded understanding of both images and videos. Unlike existing multi-...
SemComp-Bench introduces a new video generation task called Semantic Task Completion, where a model must produce a video that achieves a specified outcome while maintaining semantic alignment with a reference image. The benchmark includes the SemComp-Data dataset, spanning six domains, and a four-stage curation pipeline that transforms raw videos into standardized instances. Evaluation is performed via a vision‑language model that answers structured binary questions, yielding Outcome Achievement (OA) and Generation Reliability (GR) scores.
AdaVDR is an adaptive video deep research agent that selects and reflects on tool usage based on the task and the model’s capabilities. It constructs a specialized data pipeline to generate high‑quality QA pairs and uses model‑conditioned filtering to remove unnecessary tool calls. The agent is trained with supervised fine‑tuning and reinforcement learning, achieving top performance on the VDR‑EE benchmark and significant gains on VideoDR.
arXiv:2608.14835v2 Announce Type: replace Abstract: Dynamic scene graphs (DSGs) capture spatio-temporal interactions across videos as $\langle$subject, predicate, object$\rangle$ triplets, and underp...
OmniAssistBench is a new benchmark for evaluating omni-modal large language models (Omni-LLMs) as real‑time video assistants that actively guide users toward goals. The benchmark addresses the challenge of dynamic interaction paths by providing models with predefined priors from source videos, forcing them to follow the same routes as users. The dataset was constructed by reverse‑engineering existing Internet videos into multi‑turn clips, a process that required over 1,000 expert person‑hours. Results show that proprietary Gemini‑3‑Pro scores 66.4/100 while open‑source Qwen3‑Omni‑Instruct scores 51.2, revealing that current models often give incorrect or incomplete answers, struggle with visual prompts, and fail to maintain context or delay responses until target events.
Video-IFBench is a new benchmark designed to evaluate how well multimodal large language models (MLLMs) follow user-specified instructions in video understanding tasks. It introduces an instruction taxonomy with four templates—single-task, multi-task, selection, and nested—covering 32 task types and 39 constraint categories that span semantic and format requirements. The benchmark was built using a semi-automatic pipeline that combines MLLMs, programmatic processing, and human verification, producing 1.5K samples, and a large-scale evaluation of over 20 recent MLLMs shows that instruction following remains difficult, especially for complex constraints and conditional structures.
arXiv:2609.37407v1 Announce Type: new Abstract: While recent video foundation models excel at generating high-quality short videos, long-form video generation remains a critical challenge, where a ma...
arXiv:2604.12335v2 Announce Type: replace-cross Abstract: Training multimodal large language models (MLLMs) for video understanding requires large-scale annotated data spanning diverse tasks such as...
arXiv:2608.23329v1 Announce Type: cross Abstract: Open-world video understanding often requires a model to locate sparse visual evidence and acquire external knowledge that is absent from the video a...