All updates

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…

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.

In this episode, we test MiniMax H3 in ComfyUI with 25 free workflows, plus 2 bonus workflows. H3 is a video model that makes video with sound, from your images, videos and audio. We test them all, and see what H3 does well and where it struggles. Part 1, change your video: a…

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…

Three MiniMax H3 models, three prices. Here's when to use H3 Highspeed, H3 Lite, and H3.

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…

PLANET HELL is an AI Sci-FI Short Film expanding on the original AI Film trailer made in Veo-3.1 This is an ongoing series about a prison planet where every inmate gets one bullet. This latest film was made with Seedance 2.5, GPT Image 2 and 2.5, Nano Banana and Midjourney, and…

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…

Today I am testing the most popular AI dictation apps side by side: Wispr Flow, Superwhisper, OpenWhispr, and the OpenWispr Android app from Team 3MB. These tools promise to turn your voice into clean, ready-to-send text in any app, but they work very differently. Some run in…

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.

TaoMate-HD is an autoregressive streaming model that generates high-resolution audiovisual content in chunks with progressive denoising, history-based continuation, and optimized parallel deployment for faster, coherent long-form output.

Recent advances in generative AI have enabled multiple pathways for high-fidelity video synthesis: text-to-video, image-to-video animation, and video outpainting. Empirical side-by-side evaluations of how untrained viewers perceive and distinguish these outputs from real footage…

Autoregressive video generation has achieved impressive results on short clips, yet generating long videos remains challenging due to error accumulation over extended horizons, where each predicted frame depends on potentially imperfect previous frames. Existing approaches…

Recent advances in video generation have enabled generative video compression to reconstruct details lost during coding using pretrained spatiotemporal priors. However, a substantial gap remains between codec-induced degradation and the perturbation trajectories learned by…

Cool Shirts: https://www.qwertee.com/shop/tees?sort=price DISCOUNT CODE: Olivio Support me on Patreon & get my WORKFLOWS: https://www.patreon.com/oliviotutorials Get 1€ Discont with Code "Olivio" on https://www.qwertee.com/ Joint my Discord Group: https://discord.gg/XKAk7GUzAW…

Diffusion Transformer (DiT)-based text-to-image models have recently achieved remarkable success in image synthesis. However, when directly applied beyond their training resolution, these models often suffer from severe semantic degradation, where meaningful semantic content…

The scarcity of diverse, high-quality medical images remains a significant bottleneck in advancing AI-driven medical diagnostics and treatment planning. Traditional datasets often suffer from class imbalances, limited disease coverage, and insufficient variations in imaging…

Video inpainting for object removal is scored over the whole hole, although the background behind the object is visible in other frames of the shot for some hole pixels and in no frame for others. A whole-hole score therefore mixes copying with generation and cannot show which…

Visually realistic video generation does not guarantee physically consistent motion. Generated objects may exhibit implausible collision responses or rebounds without energy dissipation, while even physically valid motion predictions may not be faithfully realized during…