Collaborating with visionaries like Tony Shyu @aineuwave shows exactly what happens when creative intuition meets the power of Topview. Check out this incredible movie he crafted—taking production value to the next level with AI. #AI# #VideoGeneration# #Topview# #CreatorEconomy#

TL;DR: ATAS is a world-state-centric framework that helps AI video generators keep "the same place" and "the same event" consistent across multiple separate shots. Title: Any Time, Any Space: A World-State-Centric Framework for Agentic Video Generation URL: Key points 🗺️ Keeps scene memory (static spatial structure) and event memory (dynamic happenings) as two separate, persistent stores 🎬 Recursively compiles a screenplay into shot requests tagged with location, camera, and time ✅ Only validated observations update the world state, so a failed shot's errors don't propagate into later ones 📈 Achieves the top aggregate score across four different video-generation backbones 🧪 Hits a PSNR-C of 15.968 on spatial revisits and an event-identity score of 0.750 on recapture tasks 🙋 Wins highest human ratings on all six criteria, including 4.40/5.0 for script faithfulness and 4.22/5.0 for character consistency I appreciate that the paper is upfront about the trade-off: per-story runtime grows roughly 1.25x in exchange for that consistency. #VideoGeneration# #AIAgents#

TL;DR: A survey that treats video as an incomplete projection of a hidden world and organizes video AI research into four operations: Reading, Writing, Sharing, and Interacting. Title: Reading, Writing, Sharing, and Interacting with the World through Video URL: Key points ・👁️ Reading: infer semantics, space, time, physics, and causality from video ・🎬 Writing: generate video from text, images, camera paths, and other conditions ・🔗 Sharing: unify understanding and generation via shared tokens and objectives ・🕹️ Interacting: closed-loop world models that predict state transitions under actions ・🎞️ Temporal coherence (neighboring frames agree) is not world consistency (compatible with an admissible world trajectory) ・🪜 An evidence ladder grades claims from plausible output up to decision utility The shift in emphasis from visual fidelity to persistent state and physical and causal validity is a useful way to read the field. #WorldModels# #VideoGeneration#

🚀 Introducing Hy-Nomas: An Open Source Free-Form Interactive Video World Model Key Features: * Free-form interaction with text prompts and camera trajectories * Long-horizon generation with world memory * Lightweight LoRA adaptation for specialized domains such as robotics * Optimized for Inference on Consumer GPUs Test it and drop feedback anytime！ 🔗 Website: 💻 GitHub: 🤗 Hugging Face: 🎥 Demo: #Hy##WorldModels# #VideoGeneration# #GenerativeAI#

TL;DR: ATAS is a world-state-centric framework that helps AI video generators keep "the same place" and "the same event" consistent across multiple separate shots. Title: Any Time, Any Space: A World-State-Centric Framework for Agentic Video Generation URL: Key points 🗺️ Keeps scene memory (static spatial structure) and event memory (dynamic happenings) as two separate, persistent stores 🎬 Recursively compiles a screenplay into shot requests tagged with location, camera, and time ✅ Only validated observations update the world state, so a failed shot's errors don't propagate into later ones 📈 Achieves the top aggregate score across four different video-generation backbones 🧪 Hits a PSNR-C of 15.968 on spatial revisits and an event-identity score of 0.750 on recapture tasks 🙋 Wins highest human ratings on all six criteria, including 4.40/5.0 for script faithfulness and 4.22/5.0 for character consistency I appreciate that the paper is upfront about the trade-off: per-story runtime grows roughly 1.25x in exchange for that consistency. #VideoGeneration# #AIAgents#

🌍 TL;DR: A training-free way to re-explore a recorded video from new camera angles just dropped. It handles viewpoint changes, hole-filling, and recovering earlier appearances entirely through a frozen model's own attention. Title: World in World: Explore the World with World Models URL: Points 🎥 Four evidence sources (source video, target-view scene, rendered geometry, retrieved history) get converted into clean K/V fed straight into the frozen model 🧭 CGAR routes queries toward reliable source-video tokens using persistent point correspondence and camera geometry 🎚 EWA independently tunes each auxiliary channel's contribution using existing attention outputs, with zero extra network evaluations 🏆 Top VBench score of 85.192 and lowest rotation error (2.8326°) against 6 baselines including ReCamMaster 🔧 Same framework also powers bullet-time rendering, video stabilization, editing, and motion transfer Getting this much versatility with zero retraining feels like a win for evidence engineering over model adaptation. #WorldModels# #VideoGeneration#

Can you generate a video that spins a product or person a full 360° and still keeps it "the same thing"? 🔄 OmniShowcase: Omnidirectional Subject-Consistent Video Generation from Multi-View Images It generates subject-consistent orbit videos from four multi-view reference images. 🧱 Highlight 1: auto-synthesized 3D-consistent data From one 3D mesh, it renders the target orbit and four reference views at 90° intervals. Backgrounds come from video inpainting, with a VLM gating every stage. Curation costs about $120 in total, and the roughly 4,000-video Showcase-4K is planned for release. 📐 Highlight 2: normal maps as in-context input It builds 3D Gaussians from the four references and feeds a normal-map turntable to Wan2.1 via LoRA. No frame-level alignment with the target is needed, and going from one to four references lifts DINO from 0.702 to 0.799. 📊 Highlight 3: a wide margin on 3D metrics On ShowcaseBench versus 3DreamBooth, GT-CD improves from 0.1717 to 0.0824 and MEt3R from 0.325 to 0.190. In the user study it won 70.9–79.4% of votes on each criterion. The takeaway: explicit 3D geometry beats adding more 2D references. Non-rigid subjects remain a weakness. #VideoGeneration# #ComputerVision#

A world model you steer with text and camera controls, with a design that even runs on a single RTX 4090 🎮 Hy-Nomas: Towards Free-form Interactive World Generation An 8B autoregressive video world model from a Tencent team. One model handles free-form text, continuous camera control, and multi-turn memory. Three highlights: 🧠 Action-aware causal attention Each segment's text is split into a description that later segments can see and actions confined to their own time window. The aim is to curb action-order errors and repeated actions across segments. ⚡ 4-step distillation and light inference Sampling drops from 100 to 4 function evaluations, combined with INT8 and sparse attention. A 100-second video (1280x704, 24 FPS) is generated at a peak of 20.53 GiB on an RTX 4090. 📊 Strong lead on text control In human evaluation of prompt following, it wins 75.0% and loses 8.3% against LingBot-World v2. Spatial memory, however, still trails Genie 3. A good read for both the design and evaluation of interactive world generation. #WorldModels# #VideoGeneration#

Physical information is sitting inside video embeddings, but models can't quite pull it out. This paper tests that gap with 8,000 controlled physics simulations. Title: World Embedding Benchmark URL: Built from 80 families across fluid mechanics, solid mechanics, dynamics, and optics, this benchmark measures how much physics video embeddings actually "understand." Three highlights stand out. 🔍 Physical information exists, but is hard to extract Pre-trained models show weak retrieval (Recall@10 of just 0.6-7.3%), yet V-JEPA2 achieves strong physical-property regression (nRMSE ~0.94-1.02) despite having no text alignment at all. The physics is encoded — it's just not well connected to language. ⚖️ A tradeoff between alignment and recoverability Physics-specific contrastive training improves retrieval and pair classification, but actually hurts physical-property regression. Reorganizing the representation space for similarity matching comes at the cost of preserving quantitative physical information. 🎬 A real payoff in retrieval-augmented generation Using a physics-adapted model to retrieve reference videos lifted the average generation score from 0.60 to 0.66, with the biggest fidelity gains in optics and thermal domains. The takeaway for world-model design: alignment and recoverability need to be evaluated together, not optimized one at a time. #WorldModels# #VideoGeneration#

The longer a video generation gets, the harder it is to know which past frames to remember. This paper picks them based on future usefulness instead of present similarity. Title: FrameMorrow: Future-guided Frame Selection with Prospective Tokens for Long-Horizon Video Generation URL: 📝 Overview FrameMorrow selects historical frames based on relevance to future information needs rather than similarity to the present, using "prospective tokens" that predict what will matter next without actually generating future video. ❗ Problem it solves Existing methods keep frames similar to recent content, but that's myopic — visually similar frames often don't help future generation, and long videos lose coherence as a result. 🛠️ Method A frozen visual encoder (DINOv2) feeds a small set of learnable projections and attention layers that autoregressively predict prospective tokens, trained via "future-grounded ranking distillation" against actual continuation frames. Because the generator itself stays frozen and FrameMorrow just hands it explicit selected frames through a per-backbone adapter, it plugs into open-source models as well as closed-source APIs like Seedance 2.0 and Kling O3. 📊 Results Tested across 5 benchmarks and 11 models, it lifts interactive-generation consistency from 84.95 to 89.30 and improves action alignment on the WorldMem model from 0.866 to 0.889. In a user study, raters preferred FrameMorrow's consistency 62.7% of the time versus 16.7% for native models. #VideoGeneration# #AIResearch#