Composed Video Retrieval (CoVR) retrieves a target video given a query video and a modification text describing the intended change. Existing CoVR benchmarks emphasize appearance shifts or coarse event changes and therefore do not test the ability to capture subtle, fast-paced temporal differences. We introduce TFCoVR, the first large-scale benchmark dedicated to temporally fine-grained CoVR. TF-CoVR focuses on gymnastics and diving, and provides 180K triplets drawn from FineGym and FineDiving datasets. Previous CoVR benchmarks, focusing on temporal aspect, link each query to a single target segment taken from the same video, limiting practical usefulness. In TF-CoVR, we instead construct each pair by prompting an LLM with the label differences between clips drawn from different videos; every pair is thus associated with multiple valid target videos (3.9 on average), reflecting real-world tasks such as sports-highlight generation. To model these temporal dynamics, we propose TF-CoVR-Base, a concise two-stage training framework: (i) pre-train a video encoder on fine-grained action classification to obtain temporally discriminative embeddings; (ii) align the composed query with candidate videos using contrastive learning. We conduct the first comprehensive study of image, video, and general multimodal embedding (GME) models on temporally fine-grained composed retrieval in both zero-shot and fine-tuning regimes. On TF-CoVR, TF-CoVR-Base improves zero-shot mAP@50 from 5.92 (LanguageBind) to 7.51, and after fine-tuning raises the state-of-the-art from 19.83 to 27.22.

W e introduce a novel pipeline for joint audio-visual editing that enhances the coherence between edited video and its accompanying audio. Our approach first applies state-of-the-art video editing techniques to produce the target video, then performs audio editing to align with the visual changes. T o achieve this, we present a new video-to-audio generation model that conditions on the source audio, target video, and a text prompt. W e extend the model architecture to incorporate conditional audio input and propose a data augmentation strategy that improves training efficiency. Furthermore, our model dynamically adjusts the influence of the source audio based on the complexity of the edits, preserving the original audio structure where possible. Experimental results demonstrate that our method outperforms existing approaches in maintaining audio-visual alignment and content integrity.

Unified, generalizable semantic control in video generation remains a critical open challenge. Existing methods either introduce artifacts by enforcing inappropriate pixel-wise priors from structure-based controls, or rely on non-generalizable, condition-specific finetuning or task-specific architectures. We introduce Video-As-Prompt (VAP), a new paradigm that reframes this problem as in-context generation. VAP leverages a reference video as a direct semantic prompt, guiding a frozen Video Diffusion Transformer (DiT) via a plug-and-play Mixture-of-Transformers (MoT) expert. This architecture prevents catastrophic forgetting and is guided by a temporally biased position embedding that eliminates spurious mapping priors for robust context retrieval. To power this approach and catalyze future research, we built VAP-Data, the largest dataset for semantic-controlled video generation with over 100K paired videos across 100 semantic conditions. As a single unified model, VAP sets a new state-of-the-art for open-source methods, achieving a 38.7% user preference rate that rivals leading condition-specific commercial models. VAP's strong zero-shot generalization and support for various downstream applications mark a significant advance toward general-purpose, controllable video generation.

The security risks of AI-driven video editing have garnered significant attention. Although recent studies indicate that adding perturbations to images can protect them from malicious edits, directly applying image-based methods to perturb each frame in a video becomes ineffective, as video editing techniques leverage the consistency of inter-frame information to restore individually perturbed content. To address this challenge, we leverage the temporal consistency of video content to propose a straightforward and efficient, yet highly effective and broadly applicable approach, Universal Video Consistency Guard (UVCG). UVCG embeds the content of another video(target video) within a protected video by introducing continuous, imperceptible perturbations which has the ability to force the encoder of editing models to map continuous inputs to misaligned continuous outputs, thereby inhibiting the generation of videos consistent with the intended textual prompts. Additionally leveraging similarity in perturbations between adjacent frames, we improve the computational efficiency of perturbation generation by employing a perturbation-reuse strategy. We applied UVCG across various versions of Latent Diffusion Models (LDM) and assessed its effectiveness and generalizability across multiple LDM-based editing pipelines. The results confirm the effectiveness, transferability, and efficiency of our approach in safeguarding video content from unauthorized modifications.

The rapid expansion of multimedia content has made accurately retrieving relevant videos from large collections increasingly challenging. Recent advancements in text-video retrieval have focused on cross-modal interactions, large-scale foundation model training, and probabilistic modeling, yet often neglect the crucial user perspective, leading to discrepancies between user queries and the content retrieved. To address this, we introduce MERLIN (Multimodal Embedding Refinement via LLM-based Iterative Navigation), a novel, training-free pipeline that leverages Large Language Models (LLMs) for iterative feedback learning. MERLIN refines query embeddings from a user perspective, enhancing alignment between queries and video content through a dynamic question answering process. Experimental results on datasets like MSR-VTT, MSVD, and ActivityNet demonstrate that MERLIN substantially improves Recall@1, outperforming existing systems and confirming the benefits of integrating LLMs into multimodal retrieval systems for more responsive and context-aware multimedia retrieval.

Diffusion models have achieved significant success in image and video generation. This motivates a growing interest in video editing tasks, where videos are edited according to provided text descriptions. However, most existing approaches only focus on video editing for short clips and rely on time-consuming tuning or inference. We are the first to propose Video Instruction Diffusion (VIDiff), a unified foundation model designed for a wide range of video tasks. These tasks encompass both understanding tasks (such as language-guided video object segmentation) and generative tasks (video editing and enhancement). Our model can edit and translate the desired results within seconds based on user instructions. Moreover, we design an iterative auto-regressive method to ensure consistency in editing and enhancing long videos. We provide convincing generative results for diverse input videos and written instructions, both qualitatively and quantitatively. More examples can be found at our website https://ChenHsing.github.io/VIDiff.

This paper describes the MediaEval 2020 \textit{Predicting Media Memorability} task. After first being proposed at MediaEval 2018, the Predicting Media Memorability task is in its 3rd edition this year, as the prediction of short-term and long-term video memorability (VM) remains a challenging task. In 2020, the format remained the same as in previous editions. This year the videos are a subset of the TRECVid 2019 Video-to-Text dataset, containing more action rich video content as compared with the 2019 task. In this paper a description of some aspects of this task is provided, including its main characteristics, a description of the collection, the ground truth dataset, evaluation metrics and the requirements for participants' run submissions.

In a few short years, neural-network-powered automated face swaps have gone from being mildly convincing to eerily believable. But through new research from Disney, neural face-swapping is poised to become a legitimate and high-quality tool for visual effects studios working on Hollywood blockbusters. One of the bigger challenges of creating deepfake videos, as they've come to be known, is creating a vast database of facial images of a person--thousands of different expressions and poses--that can be swapped into a target video. The larger the database and the higher the quality of the images, the better the face swaps will turn out. But the images (which are more often than not headshots of famous people) are usually pulled from sources with limited resolution.

Shelly Duvall is hiding from her crazed husband as he chops down the door with an axe. Jim Carrey sticks his head through the opening and cackles the iconic line: "Here's Johnny!" What you're seeing is not a Hollywood special effect. It wasn't done with After Effects, green screen, or with costuming and makeup. The video is a fake created by deep learning artificial intelligence – a deepfake.

In the latest example of deepfake technology, researchers have shown off new software that uses machine learning to let users edit the text transcript of a video to add, delete, or change the words coming right out of somebody's mouth. The work was done by scientists from Stanford University, the Max Planck Institute for Informatics, Princeton University, and Adobe Research, and shows that our ability to edit what people say in videos and create realistic fakes is becoming easier every day. You can see a number of examples of the system's output below, including an edited version of a famous quotation from Apocalypse Now, with the line "I love the smell of napalm in the morning" changed to "I love the smell of french toast in the morning." This work is just at the research stage right now and isn't available as consumer software, but it probably won't be long until similar services go public. Adobe, for example, has already shared details on prototype software named VoCo, which lets users edit recordings of speech as easily as a picture, and which was used in this research.