arXiv:2608.31106v1 Announce Type: new Abstract: Recent video generators often omit audio or synthesize it in a separate stage, limiting reciprocal modeling of visual dynamics and acoustic events. We...
Avatar‑Forever introduces a decoupled parallel training framework that separates efficient generation from long‑horizon robustness, allowing each capability to be trained independently. The method employs full‑parameter distillation for high‑quality generation and a lightweight long‑horizon adapter trained via Recovery‑oriented Rollout Training (RRT) to mitigate autoregressive errors. Additionally, a ForeverCache feature‑caching mechanism reduces redundant computation during streaming inference, enabling real‑time, high‑resolution avatar generation at 27.2 FPS on a single H100 GPU.
Encore is a new framework for generating long, synchronized audio‑video content. It splits the problem into local continuity, handled by iterative chunk‑wise synthesis with cross‑chunk context, and global consistency, enforced through reference audio‑video signals with shifted position embeddings. The Adaptive Signal Routing mechanism learns attention biases and residual scales to modulate the influence of each conditioning signal, enabling end‑to‑end joint audio‑video generation and infinite‑length inference.
arXiv:2610.17552v1 Announce Type: cross Abstract: Recent video super-resolution (VSR) models achieve efficient one-step generation through diffusion distillation. However, high-resolution inference r...
arXiv:2511. 18050v1 Announce Type: cross Abstract: Diffusion transformers have recently delivered strong text-to-image generation around 1K resolution, but we show that extending them to native 4K across diverse aspect ratios exposes a tightly coupled failure mode spanning positional encoding, VAE compression, and optimization.
Video DeltaNet (VDN) introduces a hybrid attention mechanism for livestream video generation, combining local Softmax attention with a bidirectional linear memory branch called Video Delta Attention (VDA). VDA updates memory once per frame, integrating spatial tokens, while separate output projections and learnable gates balance the two branches. Applied to MiniMax H3, VDN achieves a 14.5× speedup over the dense baseline, completing 14.3‑second, 768p video denoising in 6.70 seconds on eight NVIDIA B200 GPUs.
arXiv:2506. 10915v2 Announce Type: replace-cross Abstract: Text-to-video generation has significantly enriched content creation and holds the potential to evolve into powerful world simulators.
X-AVDD is a cross‑attention‑based audio‑visual dataset distillation method that explicitly models inter‑modal relationships during synthesis. By decoupling cross‑modal alignment from joint optimization, it reduces memory usage and end‑to‑end GPU‑hour cost by about eight times while producing highly aligned synthetic data. Empirical results show significant performance gains over prior methods, improving top‑1 test accuracy on VGGSound from 8.2 % to 24.0 % at 10 samples‑per‑class and enabling efficient scaling to complex architectures such as AudioCLIP.
arXiv:2508.15774v2 Announce Type: replace Abstract: Visual diffusion models achieve remarkable progress, yet they are typically trained at limited resolutions due to the lack of high-resolution data...
FlashVSR+ is a new video super‑resolution model that tackles two main issues: the high cost of high‑resolution inference and the difficulty of aligning distilled models with human preferences. It introduces a hybrid attention architecture that combines strong local processing with sparse global interactions, achieving up to twice the speed of DiT and 45 % lower peak memory, which allows untiled 4K VSR on a single 24 GB RTX 4090. For training, the authors propose Reward‑Yield Distribution Matching Distillation (Ry‑DMD), a method that uses reward‑weighted teacher rollouts to incorporate black‑box rewards into distribution matching without backpropagating through the reward model, and demonstrate that Ry‑DMD generalizes beyond VSR to text‑to‑image generation.
PRISM‑Bench is an audio‑centric diagnostic benchmark for text‑to‑audio‑video generation, built from 900 human‑verified samples. It evaluates audio along two axes—audio type (speech, music, sound) and sound‑source visibility (on‑screen vs. off‑screen)—across four perceptual dimensions (audio‑visual coherence, audio quality, audio expressiveness, and prompt following) using 35 fine‑grained criteria. The benchmark employs an enhanced MLLM‑as‑a‑Judge protocol that aligns strongly with human raters, revealing a performance gap between frontier and open‑source T2AV models and highlighting overfitting to perceptual fidelity while struggling with complex grounding and control tasks, especially for music and synchronized on‑screen audio.
Large-scale visual generators are increasingly capable but costly to train, fine-tune, and deploy. We introduce Mage-Flow, a compact 4B-scale generative stack for efficient text-to-image generation and instruction-based image editing.