#genai
- 17 Aug 2026 · Paper summary
Paper summary: Can We Defend Against AI-Generated Video Attacks on Real-World Crisis Events? A Systematic Evaluation of Detectors, Generators and Social Dissemination
RA-Bench pairs 1,830 real crisis videos with 16,056 generated clips from nine generators, and finds that no family of AI-video detectors generalises consistently, including when videos are re-shared on social media.
- 10 Jul 2026 · Paper summary
Paper summary: Vidu S1: A Real-Time Interactive Video Generation Model
Vidu S1 generates infinite-length interactive video of digital characters in real time, controlled by voice at any moment, at 540p and up to 42 FPS on consumer GPUs.
- 22 Apr 2026 · Paper summary
Paper summary: Tstars-Tryon 1.0: Robust and Realistic Virtual Try-On for Diverse Fashion Items
Tstars-Tryon 1.0 is a commercial-scale virtual try-on system built for in-the-wild photos, multi-item outfits (up to 6 references across 8 categories) and near real-time speed, released with a new benchmark.
- 16 Apr 2026 · Paper summary
Paper summary: Seedance 2.0: Advancing Video Generation for World Complexity
Seedance 2.0 is ByteDance's joint audio–video generation model accepting text, image, audio and video references, generating 4–15 second clips at 480p or 720p.
- 5 Mar 2026 · AI updates
Modular Diffusers: diffusion pipelines from composable blocks
Hugging Face Diffusers introduced ModularPipeline, which builds diffusion pipelines from reusable blocks (text encoding, image encoding, denoising, decoding) that can be inspected, swapped and run on their own.
- 19 Dec 2025 · Paper summary
Paper summary: Kling-Omni Technical Report
Kling-Omni is one end-to-end framework for video generation, editing and reasoning-based editing from mixed inputs (text, reference images, video context), built on a dedicated data system.
- 10 Dec 2025 · Paper summary
Paper summary: Wan-Move: Motion-controllable Video Generation via Latent Trajectory Guidance
Wan-Move adds precise motion control to an image-to-video model by turning dense point trajectories into a motion-aware latent condition, with no architecture change; user studies rate it close to Kling 1.5 Pro's Motion Brush.
- 20 Nov 2025 · AI updates
Nano Banana Pro (Gemini 3 Pro Image) for developers
Google DeepMind released Nano Banana Pro, an image generation and editing model built on Gemini 3 Pro, with more accurate text rendering, Google Search grounding, fine camera and lighting control, and up to 4K output.
- 20 Nov 2025 · Paper summary
Paper summary: Kandinsky 5.0: A Family of Foundation Models for Image and Video Generation
Kandinsky 5.0 is an open family of image and 10-second video generation models (6B Image Lite, 2B Video Lite, 19B Video Pro), with the full data-curation and multi-stage training pipeline described.
- 5 Aug 2025 · Paper summary
Paper summary: Qwen-Image Technical Report
Qwen-Image is an image-generation foundation model focused on rendering complex text (including Chinese) inside images and on consistent image editing, using a text-rendering curriculum and dual encoding of the input image.
- 18 Oct 2024 · Paper summary
Paper summary: Movie Gen: A Cast of Media Foundation Models
Meta's Movie Gen is a set of foundation models for 1080p HD video with synchronised audio, plus personalisation and instruction-based editing; the largest video model is a 30B Transformer.
- 30 Sept 2024 · Paper summary
Paper summary: Emu3: Next-Token Prediction is All You Need
Emu3 tokenises images, text and video into one discrete space and trains a single Transformer from scratch with only next-token prediction, reported to beat SDXL at generation and LLaVA-1.6 at perception.
- 5 Sept 2024 · Paper summary
Paper summary: Loopy: Taming Audio-Driven Portrait Avatar with Long-Term Motion Dependency
Loopy is a video diffusion model that animates a portrait from audio alone, using long-term motion modules instead of the spatial motion templates other methods add to keep movement stable.
- 21 May 2024 · Paper summary
Paper summary: FIFO-Diffusion: Generating Infinite Videos from Text without Training
FIFO-Diffusion generates arbitrarily long videos from a pretrained text-to-video diffusion model, with no extra training, by denoising a queue of frames at increasing noise levels.
- 17 May 2024 · Paper summary
Paper summary: Chameleon: Mixed-Modal Early-Fusion Foundation Models
Chameleon is one token-based model that reads and generates images and text in any order, using early fusion, plus a training recipe designed to keep this mixed-modal setup stable.
- 24 Jan 2024 · Paper summary
Paper summary: Lumiere: A Space-Time Diffusion Model for Video Generation
Lumiere generates the full duration of a video in one pass with a Space-Time U-Net that downsamples in both space and time, instead of generating keyframes and then filling in between.
- 10 Nov 2023 · Paper summary
Paper summary: LCM-LoRA: A Universal Stable-Diffusion Acceleration Module
LCM-LoRA packages latent-consistency distillation as a LoRA that plugs into Stable Diffusion models and their fine-tunes to generate images in very few steps without extra training.
- 26 Oct 2023 · Paper summary
Paper summary: A Picture is Worth a Thousand Words: Principled Recaptioning Improves Image Generation
RECAP relabels a text-to-image training set with captions from a specialised captioning model instead of web alt-text, and reports better image quality and much better prompt following.
- 6 Oct 2023 · Paper summary
Paper summary: Kandinsky: an Improved Text-to-Image Synthesis with Image Prior and Latent Diffusion
Kandinsky combines a separately trained image prior, which maps CLIP text embeddings to CLIP image embeddings, with latent diffusion and a modified MoVQ autoencoder: 3.3B parameters, open weights.
- 13 Sept 2023 · AI updates
Würstchen: diffusion in a 42× compressed latent space
Würstchen is a text-to-image diffusion model whose text-conditional stage works in a latent space compressed 42× spatially, making training and inference much cheaper than Stable Diffusion.
- 1 Sept 2023 · Paper summary
Paper summary: MVDream: Multi-view Diffusion for 3D Generation
MVDream trains a diffusion model to generate several consistent views of an object at once, then uses it as a 3D prior for text-to-3D, which reduces problems like the "two-faced horse".