All updates

Manim supports precise, programmatic video generation, but a successfully rendered educational animation can still have visual defects, missing explanations, or incorrect content. Rendering does not directly identify the code changes needed to correct these visual failures, so…

Text-to-image (T2I) systems serve users across languages and cultural contexts. However, evaluations of cultural representation focus mainly on recognizable elements such as landmarks, clothing, and activities, overlooking the cultural values that generated images convey. To…

EDGAR is a conditional diffusion framework with a fusion block and Mixture-of-Experts that refines degraded Gaussian head avatars by injecting self-supervised identity features, using refined multi-view renderings as pseudo-targets to optimize a single avatar.

Large-scale identity-labeled face datasets are essential for training modern face recognition models, yet collecting and annotating such data is costly and raises privacy and ethical concerns. Existing synthetic face generation methods typically derive identities from predefined…

Text-guided image editing with diffusion models struggles with spatial edits (e.g. manipulating objects, changing viewpoints, manipulating occlusions) that depend on the 3D structure understanding of a scene. Marigold and follow-up work demonstrate that a pretrained diffusion…

An image editing service records which user requests each edit, but the edited image itself does not contain this account information. Once the image is downloaded and shared, the service cannot identify the requesting user from the image alone. Embedding a user-specific…

Few-step autoregressive (AR) video diffusion models generate videos chunk by chunk and have made efficient video generation practical. However, further accelerating these models toward real-time generation remains challenging, as few-step distillation leaves little cross-step…

We present \textbf{WeLM-Image}, a unified image foundation model that integrates multimodal understanding into text-to-image generation, instruction-guided editing, and multi-reference synthesis. Designed for single-GPU deployment in WeChat's large-scale consumer applications…

Reinforcement learning has become an effective approach for improving video generation, but its success depends heavily on reliable reward models. Existing methods either map videos directly to opaque scalar scores or generate free-form rationales that are not explicitly…

Diffusion models for video generation have achieved remarkable success, yet precise motion control under explicit user guidance remains challenging. Recent noise-warping methods introduce temporal correlations by warping the noise according to the user-provided motion signal…

We present an automated method for auditing text-to-image generators for representational bias against ground-truth demographic benchmarks. Our approach adapts the racial proxy variables used in U.S. antidiscrimination law to provide a fine-grained comparative analysis of…

Skin lesion segmentation without annotated masks requires the target to be specified through other cues. Existing approaches rely on dermoscopy-specific rules, annotation-derived model priors, or user-selected regions. A general text-to-image diffusion transformer offers a text…

We introduce LongBanana, to our knowledge the first expert-designed and expert-verified benchmark for long-context multi-reference image synthesis in the age of agent-driven image generation. The task requires retrieving dispersed visual evidence, rejecting distractors, binding…

Autoregressive image generation has recently emerged as a competitive visual generation paradigm, yet its generation order is often predefined or heuristic, such as next-patch prediction, random-order prediction, or one-step generation for specific scales. We observe that these…

Strong single-turn editing performance does not guarantee reliable multi-turn interaction: as edited outputs are repeatedly reused, small unintended changes can accumulate into severe visual drift. This long-term drift is often mitigated through training on multi-turn data or…

Adversarial training enables one- and few-step image and video generation, but conventional adversarial losses supervise critic values without directly specifying the gradients used to update the generator. We present NEXTGAN, which trains a scalar critic by regressing its input…

Embodied video world models predict scene changes caused by robot actions, yet generated object responses may remain inconsistent with task intent and contact dynamics. Critical responses in embodied scenes often concentrate in local regions and specific time intervals. Inspired…

Sparse attention accelerates few-step video diffusion transformers by computing a few key blocks exactly and compensating for the rest. However, existing sparse attention still faces key challenges, including identification overhead from finding critical blocks at every…

Identity-preserving video generation aims to synthesize high-quality and visually realistic videos from textual and visual conditions while faithfully maintaining the subject's identity across diverse motions. Prior methods have extensively introduced reference-guided control…

Digital human e-commerce video generation requires preserving product appearance, human identity, and overall video style during dynamic demonstrations. Although recent models support multiple reference images, their general conditioning schemes do not explicitly distinguish the…

LiveFoley improves streaming video-to-audio generation by using a trailing sliding window for visual alignment and expanded temporal denoising windows for audio quality with minimal latency.

Conversational video characters must generate video as fast as it plays, for one user at a time: at 24 fps, a sustained 166.7 ms per four-frame chunk at batch size one, with only a few chunks of buffering. We describe the system behind Runway Characters, which serves a…

Text-to-image (T2I) systems serve users across languages and cultural contexts. However, evaluations of cultural representation focus mainly on recognizable elements such as landmarks, clothing, and activities, overlooking the cultural values that generated images convey. To…

Group Relative Policy Optimization (GRPO) has advanced generative model post-training in verifiable domains such as code and mathematics, but evaluating video quality, especially emotional expression, often depends on subjective aesthetics and has no uniquely correct target…

TurnWise improves streaming video understanding by assigning turn-level credit and balancing decision weights in group relative policy optimization.

Test-time scaling (TTS) improves generative models by allocating additional inference-time computation to sampling, search, and verification. Existing image generation approaches, however, typically rank completed images or image-centric intermediate states. We study TTS for…

Video generation has progressed from isolated clips to coherent short-form audiovisual sequences, and the bottleneck is shifting from rendering to semantic control. This report summarizes the observations and insights on multimodal understanding gained from successive…

Latent diffusion and flow matching models are ubiquitous in state-of-the-art generative video models. While highly effective, relying on pretrained autoencoders to compress the data into a lower-dimensional latent space also comes with drawbacks. First, training proceeds in two…

In-image machine translation (IIMT) replaces text in an image with its translation while preserving the surrounding visual context. Existing task-specific models achieve convincing performance but require massive paired training images before and after translation. As advanced…

Diffusion-based data augmentation has emerged as a powerful strategy for addressing data-scarce image classification. Recent pipelines typically adapt a pretrained text-to-image diffusion model to a limited number of class examples and subsequently sample synthetic images using…

RoboTryOn translates human demonstration videos into robot videos to train control policies without real robot demonstrations, improving success rates and reducing data collection costs.

DreamX leverages a pretrained video diffusion model to learn compact, appearance-invariant latent actions that enable cross-embodiment transfer and efficient policy learning.

Understanding how neurons respond to visual stimulus is an important question in both deep learning and neuroscience. It has significant implications for enhancing the interpretability of black-box artificial neural networks and for understanding the visual representation in the…

Many manipulation goals are easier to visualize than to specify numerically through explicit geometric constraints. We present VerifLM, a framework that converts image-edited goal configurations into physically verified 6-DoF object poses for planning. Using an edited image…

Pose-driven human video diffusion animates a person from a single reference image and a target pose sequence, yet it struggles to preserve identity, rendering the face and garments unfaithfully as the pose departs from it. Personalization is the standard remedy, but its…

Autoregressive video diffusion models (AR-VDMs) enable real-time, long-horizon video generation but degrade rapidly once the inference horizon outruns the training horizon, due to compounding exposure bias. To stabilize long rollouts, recent methods heuristically adopt frame…

Reinforcement learning (RL) elevates joint audio-video generation, but optimizing heterogeneous, modality-specific audio and video rewards often introduces cross-modal interference. On-policy distillation (OPD) alleviates this issue by transferring the capabilities of…

Existing benchmarks for joint text-to-audio-video (T2AV) generation evaluate perceptual quality and audiovisual alignment, but balanced sound-source sam?pling and separate reporting by shot structure remain limited. This makes it dif?ficult to determine whether a specific sound…

Prompt engineering is commonly used to encourage text-to-image models to generate more detailed images. However, it does not provide reliable, continuous, or disentangled control. We analyze feature tokens produced with and without detail-enhancing prompts across six diffusion…

VidDub is a two-step diffusion video dubbing model that uses replacement-aware cross-audio supervision to avoid conflicting source-mouth targets and achieves strong lip synchronization with high visual fidelity.