Video restoration (VR) aims to recover high-quality videos from degraded ones. Although recent zero-shot VR methods using pre-trained diffusion models (DMs) show good promise, they suffer from approximation errors during reverse diffusion and insufficient temporal consistency. Moreover, dealing with 3D video data, VR is inherently computationally intensive. In this paper, we advocate viewing the reverse process in DMs as a function and present a novel Maximum a Posterior (MAP) framework that directly parameterizes video frames in the seed space of DMs, eliminating approximation errors. We also introduce strategies to promote bilevel temporal consistency: semantic consistency by leveraging clustering structures in the seed space, and pixel-level consistency by progressive warping with optical flow refinements. Extensive experiments on multiple virtual reality tasks demonstrate superior visual quality and temporal consistency achieved by our method compared to the state-of-the-art.

Yilun Zhu, Yuan Zhuang, Nikhita Vedula, Dushyanta Dhyani, Shaoyuan Xu, Mohsen Bayati, Bryan Wang, Shervin Malmasi. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

Music shapes the tone of videos, yet creators find it hard to find soundtracks that match their video’s mood and narrative. Recent text-to-music models let creators generate music from text prompts, but our formative study (N=8) shows creators struggle to construct diverse prompts, quickly review and compare tracks, and understand their impact on the video. We present VidTune, a system that supports soundtrack creation by generating diverse music options from a creator’s prompt and producing contextual thumbnails for rapid review. VidTune extracts representative video subjects to ground thumbnails in context, maps each track’s valence and energy onto visual cues like color and brightness, and depicts prominent genres and instruments. Creators can refine tracks with natural language edits, which VidTune expands into new generations. In a controlled user study (N=12) and an exploratory case study (N=6), participants found VidTune helpful for efficiently reviewing and comparing music options and described the process as playful and enriching.

Video storytelling is often constrained by available material, limiting creative expression and leaving undesired narrative gaps. Generative video offers a new way to address these limitations by augmenting captured media with tailored visuals. To explore this potential, we interviewed eight video creators to identify opportunities and challenges in integrating generative video into their workflows. Building on these insights and established filmmaking principles, we developed Vidmento, a tool for authoring hybrid video stories that combine captured and generated media through context-aware expansion. Vidmento surfaces opportunities for story development, generates clips that blend stylistically and narratively with surrounding media, and provides controls for refinement. In a study with 12 creators, Vidmento supported narrative development and exploration by systematically expanding initial materials with generative media, enabling expressive video storytelling aligned with creative intent. We highlight how creators bridge story gaps with generative content and where they find this blending capability most valuable.

Scriptwriting has traditionally been text-centric, a modality that only partially conveys the produced audiovisual experience. A formative study with professional writers informed us that connecting textual and audiovisual modalities can aid ideation and iteration, especially for writing dialogues. In this work, we present Script2Screen, an AI-assisted tool that integrates scriptwriting with audiovisual scene creation in a unified, synchronized workflow. Focusing on dialogues in scripts, Script2Screen generates expressive scenes with emotional speeches and animated characters through a novel text-to-audiovisual-scene pipeline. The user interface provides fine-grained controls, allowing writers to fine-tune audiovisual elements such as character gestures, speech emotions, and camera angles. A user study with both novice and professional writers from various domains demonstrated that Script2Screen’s interactive audiovisual generation enhances the scriptwriting process, facilitating iterative refinement while complementing - rather than replacing - their creative efforts.

Synthesizers are powerful tools that allow musicians to create dynamic and original sounds. Existing commercial interfaces for synthesizers typically require musicians to interact with complex low-level parameters or to manage large libraries of premade sounds. To address these challenges, we implement SynthScribe — a fullstack system that uses multimodal deep learning to let users express their intentions at a much higher level. We implement features which address a number of difficulties, namely 1) searching through existing sounds, 2) creating completely new sounds, and 3) making meaningful modifications to a given sound. This is achieved with three main features: a multimodal search engine for a large library of synthesizer sounds; a user centered genetic algorithm by which completely new sounds can be created and selected given the users preferences; a sound editing support feature which highlights and gives examples for key control parameters with respect to a text or audio based query. The results of our user studies show SynthScribe is capable of reliably retrieving and modifying sounds while also affording the ability to create completely new sounds that expand a musicians creative horizon.

Video creation has become increasingly popular, yet the expertise and effort required for editing often pose barriers to beginners. In this paper, we explore the integration of large language models (LLMs) into the video editing workflow to reduce these barriers. Our design vision is embodied in LAVE, a novel system that provides LLM-powered agent assistance and language-augmented editing features. LAVE automatically generates language descriptions for the user’s footage, serving as the foundation for enabling the LLM to process videos and assist in editing tasks. When the user provides editing objectives, the agent plans and executes relevant actions to fulfill them. Moreover, LAVE allows users to edit videos through either the agent or direct UI manipulation, providing flexibility and enabling manual refinement of agent actions. Our user study, which included eight participants ranging from novices to proficient editors, demonstrated LAVE’s effectiveness. The results also shed light on user perceptions of the proposed LLM-assisted editing paradigm and its impact on users’ creativity and sense of co-creation. Based on these findings, we propose design implications to inform the future development of agent-assisted content editing.

Live and pre-recorded video tutorials are an effective means for teaching physical skills such as cooking or prototyping electronics. A dedicated cameraperson following an instructor’s activities can improve production quality. However, instructors who do not have access to a cameraperson’s help often have to work within the constraints of static cameras. We present Stargazer, a novel approach for assisting with tutorial content creation with a camera robot that autonomously tracks regions of interest based on instructor actions to capture dynamic shots. Instructors can adjust the camera behaviors of Stargazer with subtle cues, including gestures and speech, allowing them to fluidly integrate camera control commands into instructional activities. Our user study with six instructors, each teaching a distinct skill, showed that participants could create dynamic tutorial videos with a diverse range of subjects, camera framing, and camera angle combinations using Stargazer.

Voice assistants have afforded users rich interaction opportunities to access information and issue commands in a variety of contexts. However, some users feel uneasy or creeped out by voice assistants, leading to a decreased desire to use them. As there has yet to be a comprehensive understanding of the factors that cause users to perceive voice assistants as being creepy, this research developed an empirical scale to measure the creepiness inherent in various voice assistants. Utilizing prior scale creation methodologies, a 7-item Perceived Creepiness of Voice Assistants Scale (PCAS) was created and validated. The scale measures how creepy a new voice assistant would be for users of voice assistants. The scale was developed to ensure that researchers and designers can evaluate the next generation of voice assistants before such voice assistants are released to the wider public.

Conversational agents show the promise to allow users to interact with mobile devices using language. However, to perform diverse UI tasks with natural language, developers typically need to create separate datasets and models for each specific task, which is expensive and effort-consuming. Recently, pre-trained large language models (LLMs) have been shown capable of generalizing to various downstream tasks when prompted with a handful of examples from the target task. This paper investigates the feasibility of enabling versatile conversational interactions with mobile UIs using a single LLM. We designed prompting techniques to adapt an LLM to mobile UIs. We experimented with four important modeling tasks that address various scenarios in conversational interaction. Our method achieved competitive performance on these challenging tasks without requiring dedicated datasets and training, offering a lightweight and generalizable approach to enable language-based mobile interaction.

On shopping websites, product images of low quality negatively affect customer experience. Although there are plenty of work in detecting images with different defects, few efforts have been dedicated to correct those defects at scale. A major challenge is that there are thousands of product types and each has specific defects, therefore building defect specific models is unscalable. In this paper, we propose a unified Image-to-Image (I2I) translation model to correct multiple defects across different product types. Our model leverages an attention mechanism which hierarchically incorporates high-level defect groups and specific defect types to guide the network to focus on defect-related image regions. Evaluated on eight public datasets, our model reduces the Frechet Inception Distance (FID) by 24.6% in average compared with MoNCE, the state-of-the-art I2I method. Another practical challenge on shopping websites is the lack of high quality paired images. We extend our model to be semi-paired by leveraging both paired and unpaired data. Tested on a shopping website dataset to correct three image defects, our model reduces (FID) by 63.2% in average compared with WS-I2I, the state-of-the art semi-paired I2I method.

Semi-supervised learning (SSL) has become a crucial approach in deep learning as a way to address the challenge of limited labeled data. The success of deep neural networks heavily relies on the availability of large-scale high-quality labeled data. However, the process of data labeling is time-consuming and unscalable, leading to shortages in labeled data. SSL aims to tackle this problem by leveraging additional unlabeled data in the training process. One of the popular SSL algorithms, FixMatch [1], trains identical weight-sharing teacher and student networks simultaneously using a siamese neural network (SNN). However, it is prone to performance degradation when the pseudo labels are heavily noisy in the early training stage. We present KD-FixMatch, a novel SSL algorithm that addresses the limitations of FixMatch by incorporating knowledge distillation. The algorithm utilizes a combination of sequential and simultaneous training of SNNs to enhance performance and reduce performance degradation. Firstly, an outer SNN is trained using labeled and unlabeled data. After that, the network of the well-trained outer SNN generates pseudo labels for the unlabeled data, from which a subset of unlabeled data with trusted pseudo labels is then carefully created through high-confidence sampling and deep embedding clustering. Finally, an inner SNN is trained with the labeled data, the unlabeled data, and the subset of unlabeled data with trusted pseudo labels. Experiments on four public data sets demonstrate that KD-FixMatch outperforms FixMatch in all cases. Our results indicate that KD-FixMatch has a better training starting point that leads to improved model performance compared to FixMatch.

Text-to-image generative models have demonstrated remarkable capabilities in generating high-quality images based on textual prompts. However, crafting prompts that accurately capture the user’s creative intent remains challenging. It often involves laborious trial-and-error procedures to ensure that the model interprets the prompts in alignment with the user’s intention. To address these challenges, we present Promptify, an interactive system that supports prompt exploration and refinement for text-to-image generative models. Promptify utilizes a suggestion engine powered by large language models to help users quickly explore and craft diverse prompts. Our interface allows users to organize the generated images flexibly, and based on their preferences, Promptify suggests potential changes to the original prompt. This feedback loop enables users to iteratively refine their prompts and enhance desired features while avoiding unwanted ones. Our user study shows that Promptify effectively facilitates the text-to-image workflow, allowing users to create visually appealing images on their first attempt while requiring significantly less cognitive load than a widely-used baseline tool.

Multi-modal learning with both text and images benefits multiple applications, such as attribute extraction for e-commerce products. In this paper, we propose Cross-Modality Attention Contrastive Language-Image Pre-training (CMA-CLIP), a new multi-modal architecture to jointly learn the fine-grained inter-modality relationship. It fuses CLIP with a sequence-wise attention module and a modality-wise attention module. The network uses CLIP to bridge the inter-modality gap at the global level, and uses the sequence-wise attention module to capture the fine-grained alignment between text and images. Besides, it leverages a modality-wise attention module to learn the relevance of each modality to downstream tasks, making the network robust against irrelevant modalities. CMA-CLIP outperforms the state-of-the-art method on Fashion-Gen by 5.5% in accuracy, achieves competitive performance on Food101 and performance on par with the state-of-the-art method on MM-IMDb. We also demonstrate CMA-CLIP’s robustness against irrelevant modalities on an Amazon dataset for the task of product attribute extraction.

Following the prevalence of short-form video, short-form voice content has emerged on social media platforms like Twitter and Facebook. A challenge that creators face is hard constraints on the content length. If the initial recording is not short enough, they need to re-record or edit their content. Both are time-consuming, and the latter, if supported, can have a learning curve. Moreover, creators need to manually create multiple versions to publish content on platforms with different length constraints. To simplify this process, we present ROPE1 (Record Once, Post Everywhere). Creators can record voice content once, and our system will automatically shorten it to all length limits by removing parts of the recording for each target. We formulate this as a combinatorial optimization problem and propose a novel algorithm that automatically selects optimal sentence combinations from the original content to comply with each length constraint. Creators can customize the algorithmically shortened content by specifying sentences to include or exclude. Our system can also use the user-specified constraints to recompute and provides a new version. We conducted a user study comparing ROPE with a sentence-based manual editing baseline. The results show that ROPE can generate high-quality edits, alleviating the cognitive loads of creators for shortening content. While our system and user study address short-form voice content specifically, we believe that the same concept can also be applied to other media such as video with narration and dialog.

Learning musical instruments using online instructional videos has become increasingly prevalent. However, pre-recorded videos lack the instantaneous feedback and personal tailoring that human tutors provide. In addition, existing video navigations are not optimized for instrument learning, making the learning experience encumbered. Guided by our formative interviews with guitar players and prior literature, we designed Soloist, a mixed-initiative learning framework that automatically generates customizable curriculums from off-the-shelf guitar video lessons. Soloist takes raw videos as input and leverages deep-learning based audio processing to extract musical information. This back-end processing is used to provide an interactive visualization to support effective video navigation and real-time feedback on the user's performance, creating a guided learning experience. We demonstrate the capabilities and specific use-cases of Soloist within the domain of learning electric guitar solos using instructional YouTube videos. A remote user study, conducted to gather feedback from guitar players, shows encouraging results as the users unanimously preferred learning with Soloist over unconverted instructional videos.

Mobile User Interface Summarization generates succinct language descriptions of mobile screens for conveying important contents and functionalities of the screen, which can be useful for many language-based application scenarios. We present Screen2Words, a novel screen summarization approach that automatically encapsulates essential information of a UI screen into a coherent language phrase. Summarizing mobile screens requires a holistic understanding of the multi-modal data of mobile UIs, including text, image, structures as well as UI semantics, motivating our multi-modal learning approach. We collected and analyzed a large-scale screen summarization dataset annotated by human workers. Our dataset contains more than 112k language summarization across ∼ 22k unique UI screens. We then experimented with a set of deep models with different configurations. Our evaluation of these models with both automatic accuracy metrics and human rating shows that our approach can generate high-quality summaries for mobile screens. We demonstrate potential use cases of Screen2Words and open-source our dataset and model to lay the foundations for further bridging language and user interfaces.

Input techniques have been drawing abiding attention along with the continual miniaturization of personal computers. In this paper, we present BlyncSync, a novel multi-modal gesture set that leverages the synchronicity of touch and blink events to augment the input vocabulary of smartwatches with a rapid gesture, while at the same time, offers a solution to the false activation problem of blink-based input. BlyncSync contributes the concept of a mutual delimiter, where two modalities are used to jointly delimit the intention of each other's input. A study shows that BlyncSync is 33% faster than using a baseline input delimiter (physical smartwatch button), with only 150ms in overhead cost compared to traditional touch events. Furthermore, our data indicates that the gesture can be tuned to elicit a true positive rate of 97% and a false positive rate of 1.68%.

Music creation is typically composed of two parts: composing the musical score, and then performing the score with instruments to make sounds. While recent work has made much progress in automatic music generation in the symbolic domain, few attempts have been made to build an AI model that can render realistic music audio from musical scores. Directly synthesizing audio with sound sample libraries often leads to mechanical and deadpan results, since musical scores do not contain performance-level information, such as subtle changes in timing and dynamics. Moreover, while the task may sound like a text-to-speech synthesis problem, there are fundamental differences since music audio has rich polyphonic sounds. To build such an AI performer, we propose in this paper a deep convolutional model that learns in an end-to-end manner the score-to-audio mapping between a symbolic representation of music called the pianorolls and an audio representation of music called the spectrograms. The model consists of two subnets: the ContourNet, which uses a U-Net structure to learn the correspondence between pianorolls and spectrograms and to give an initial result; and the TextureNet, which further uses a multi-band residual network to refine the result by adding the spectral texture of overtones and timbre. We train the model to generate music clips of the violin, cello, and flute, with a dataset of moderate size. We also present the result of a user study that shows our model achieves higher mean opinion score (MOS) in naturalness and emotional expressivity than a WaveNet-based model and two off-the-shelf synthesizers. We open our source code at https://github.com/bwang514/PerformanceNet

We present in this paper PerformacnceNet, a neural network model we proposed recently to achieve score-to-audio music generation. The model learns to convert a music piece from the symbolic domain to the audio domain, assigning performance-level attributes such as changes in velocity automatically to the music and then synthesizing the audio. The model is therefore not just a neural audio synthesizer, but an AI performer that learns to interpret a musical score in its own way. The code and sample outputs of the model can be found online at https://github.com/bwang514/PerformanceNet.

Proper ergonomics improves productivity and reduces risks for injuries such as tendinosis, tension neck syndrome, and back injuries. Despite having ergonomics standards and guidelines for computer usage since the 1980s, injuries due to poor ergonomics remain widespread. We present ActiveErgo, the first active approach to improving ergonomics by combining sensing and actuation of motorized furniture. It provides automatic and personalized ergonomics of computer workspaces in accordance to the recommended ergonomics guidelines. Our prototype system uses a Microsoft Kinect sensor for skeletal sensing and monitoring to determine the ideal furniture positions for each user, then uses a combination of automatic adjustment and real-time feedback to adjust the computer monitor, desk, and chair positions. Results from our 12-person user study demonstrated that ActiveErgo significantly improves ergonomics compared to manual configuration in both speed and accuracy, and helps significantly more users to fully meet ergonomics guidelines.

The rise of Maker communities and open-source electronic prototyping platforms have made electronic circuit projects increasingly popular around the world. Although there are software tools that support the debugging and sharing of circuits, they require users to manually create the virtual circuits in software, which can be time-consuming and error-prone. We present CircuitSense, a system that automatically recognizes the wires and electronic components placed on breadboards. It uses a combination of passive sensing and active probing to detect and generate the corresponding circuit representation in software in real-time. CircuitSense bridges the gap between the physical and virtual representations of circuits. It enables users to interactively construct and experiment with physical circuits while gaining the benefits of using software tools. It also dramatically simplifies the sharing of circuit designs with online communities.

For makers and developers, circuit prototyping is an integral part of building electronic projects. Currently, it is common to build circuits based on breadboard schematics that are available on various maker and DIY websites. Some breadboard schematics are used as is without modification, and some are modified and extended to fit specific needs. In such cases, diagrams and schematics merely serve as blueprints and visual instructions, but users still must physically wire the breadboard connections, which can be time-consuming and error-prone. We present CircuitStack, a system that combines the flexibility of breadboarding with the correctness of printed circuits, for enabling rapid and extensible circuit construction. This hybrid system enables circuit reconfigurability, component reusability, and high efficiency at the early stage of prototyping development.