WorldLens
VBench
[TPAMI 2026] 3D and 4D World Modeling: A Survey
This page lists names, links and short descriptions. The original list on GitHub is the source and belongs to its authors.
VBench
(arXiv 2026): log retiming and stereo references separate driving scene response from camera following. [Project] Code pending.
(arXiv 2026): source-video exploration evaluates event progression, camera control and revisit consistency.
(arXiv 2026): online agent-selected camera actions probe long-horizon spatial exploration. Paper-linked repository not verified accessible.
(arXiv 2026): matched action sequences compare video rollouts against Unity simulator states. [Code]
(arXiv 2026): a diagnostic separating scene setup, requested-change execution, and delayed reveal in video world models. [Code] [Project]
(arXiv 2026): separate diagnostic tracks evaluate agent action selection under fixed environment responses and world-model state evolution under fixed action interventions; track scores use different tasks and metrics.
(NeurIPS 2026): a robustness benchmark for occupancy forecasting under upstream perception errors and controlled occupancy-state corruptions. [Code] [Dataset] [Project]
(arXiv 2026): a camera-centric benchmark for semantic and panoptic occupancy prediction and 4D forecasting across CARLA, nuScenes, and KITTI-360, using persistent instance identities and a shared valid-mask protocol. Code pending; dataset upload in progress. [Code] [Project]
(arXiv 2026): an exploration-and-revisit benchmark separating input preservation, generated-content persistence, and 3D self-consistency. [Code] [Project]
(ICCV 2025): a unified evaluation benchmark for 3D-, 4D-, and video-based world generation models. [Code] [Dataset] [Project]
ICLR 2026
February 4-6, 2026
ICML 2025
causal diffusion representation autoencoders for interactive video world models, with selectable spatial/temporal compression and shared diffusion decoding. [Paper] [Project]
A unified codebase for world models, providing a standardized pipeline interface over existing open-source models (Matrix-Game-2, Hunyuan-GameCraft, FlashWorld, Cosmos-Predict-2.5, and others) across video generation, 3D scene generation, and reasoning. Apache-2.0.
A structured knowledge hub for AI world models, with model profiles, research syntheses, comparisons, benchmarks, and a practical taxonomy spanning video, 3D/4D, occupancy, LiDAR, robotics, and autonomous driving. Leaderboard
hesreallyhim/awesome-claude-code
A hand-picked collection of the finest of resources for the most awesome of agents, Claude Code, the undisputed champion of coding companions, from the unstoppable team…
VoltAgent/awesome-agent-skills
A curated collection of 1000+ agent skills from official dev teams and the community, compatible with Claude Code, Codex, Gemini CLI, Cursor, and more.
EthicalML/awesome-production-machine-learning
A curated list of awesome open source libraries to deploy, monitor, version and scale your machine learning
analysis-tools-dev/static-analysis
⚙️ A curated list of static analysis (SAST) tools and linters for all programming languages, config files, build tools, and more. The focus is on tools which improve…
YouMind-OpenLab/awesome-nano-banana-pro-prompts
🍌 World's largest Nano Banana Pro prompt library — 10,000+ curated prompts with preview images, 16 languages. Google Gemini AI image generation. Free & open source.
ossu/computer-science
🎓 Path to a free self-taught education in Computer Science!