mlsanityData to 7 Oct 2026

VBench-2.0: Advancing Video Generation Benchmark Suite for Intrinsic Faithfulness

arXiv 2503.21755 ↗First posted 27 Mar 2025cs.CVcited by 289

On the map: World models → Video-generation world models and simulators

Abstract (arXiv)

Video generation has advanced significantly, evolving from producing unrealistic outputs to generating videos that appear visually convincing and temporally coherent. To evaluate these video generative models, benchmarks such as VBench have been developed to assess their faithfulness, measuring factors like per-frame aesthetics, temporal consistency, and basic prompt adherence. However, these aspects mainly represent superficial faithfulness, which focus on whether the video appears visually convincing rather than whether it adheres to real-world principles. While recent models perform increasingly well on these metrics, they still struggle to generate videos that are not just visually plausible but fundamentally realistic. To achieve real "world models" through video generation, the next frontier lies in intrinsic faithfulness to ensure that generated videos adhere to physical laws, commonsense reasoning, anatomical correctness, and compositional integrity. Achieving this level of realism is essential for applications such as AI-assisted filmmaking and simulated world modeling. To bridge this gap, we introduce VBench-2.0, a next-generation benchmark designed to automatically evaluate video generative models for their intrinsic faithfulness. VBench-2.0 assesses five key dimensions: Human Fidelity, Controllability, Creativity, Physics, and Commonsense, each further broken down into fine-grained capabilities. Tailored to individual dimensions, our evaluation framework integrates generalists such as SOTA VLMs and LLMs, and specialists, including anomaly detection methods proposed for video generation. We conduct extensive human annotations to ensure evaluation alignment with human judgment. By pushing beyond superficial faithfulness toward intrinsic faithfulness, VBench-2.0 aims to set a new standard for the next generation of video generative models in pursuit of intrinsic faithfulness.

Key facts. Cited by 289 papers in the map (data through 7 Oct 2026). Placed in Video-generation world models and simulators, one of 11 directions in World models. Builds on 80 of its 93 references that are in the map. First posted on arXiv 27 Mar 2025.

Builds on

80 of 93 references.

Closest ancestors · 8

Same direction, newest first

  1. 2025Wan: Open and Advanced Large-Scale Video Generative Models26 Mar
  2. 2024CogVideoX: Text-to-Video Diffusion Models with An Expert Transformer12 Aug
  3. 2023Stable Video Diffusion: Scaling Latent Video Diffusion Models to Large Datasets25 Nov
  4. 2022Scalable Diffusion Models with Transformers19 Dec
  5. 2022MAGVIT: Masked Generative Video Transformer10 Dec
  6. 2020Score-Based Generative Modeling through Stochastic Differential Equations26 Nov
  7. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale22 Oct
  8. 2020Denoising Diffusion Probabilistic Models19 Jun

Other references · 72

  1. 2023GPT-4 Technical Reportcited by 18,132
  2. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,267
  3. 2020Denoising Diffusion Implicit Modelscited by 9,655
  4. 2015Deep Unsupervised Learning using Nonequilibrium Thermodynamicscited by 7,005
  5. 2023Adding Conditional Control to Text-to-Image Diffusion Modelscited by 5,466
  6. 2025Qwen2.5-VL Technical Reportcited by 5,418
  7. 2024Qwen2.5 Technical Reportcited by 4,713
  8. 2017Neural Discrete Representation Learningcited by 4,650
  9. 2023SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesiscited by 4,290
  10. 2024Scaling Rectified Flow Transformers for High-Resolution Image Synthesiscited by 4,094
  11. 2020Taming Transformers for High-Resolution Image Synthesiscited by 3,229
  12. 2016Improved Techniques for Training GANscited by 2,938
38 more
  1. 2015ShapeNet: An Information-Rich 3D Model Repositorycited by 2,880
  2. 2020RAFT: Recurrent All-Pairs Field Transforms for Optical Flowcited by 2,385
  3. 2022Imagen Video: High Definition Video Generation with Diffusion Modelscited by 1,765
  4. 2023VBench: Comprehensive Benchmark Suite for Video Generative Modelscited by 1,755
  5. 2024HunyuanVideo: A Systematic Framework For Large Video Generative Modelscited by 1,597
  6. 2023Align your Latents: High-Resolution Video Synthesis with Latent Diffusion Modelscited by 1,465
  7. 2023AnimateDiff: Animate Your Personalized Text-to-Image Diffusion Models without Specific Tuningcited by 1,418
  8. 2023T2I-Adapter: Learning Adapters to Dig out More Controllable Ability for Text-to-Image Diffusion Modelscited by 1,339
  9. 2021SimMIM: A Simple Framework for Masked Image Modelingcited by 1,147
  10. 2022Multi-Concept Customization of Text-to-Image Diffusioncited by 1,099
  11. 2022CogVideo: Large-scale Pretraining for Text-to-Video Generation via Transformerscited by 973
  12. 2025Cosmos World Foundation Model Platform for Physical AIcited by 833
  13. 2023Objaverse-XL: A Universe of 10M+ 3D Objectscited by 767
  14. 2021CogView: Mastering Text-to-Image Generation via Transformerscited by 691
  15. 2023Text2Video-Zero: Text-to-Image Diffusion Models are Zero-Shot Video Generatorscited by 666
  16. 2023ModelScope Text-to-Video Technical Reportcited by 651
  17. 2024Movie Gen: A Cast of Media Foundation Modelscited by 614
  18. 2024Thinking in Space: How Multimodal Large Language Models See, Remember, and Recall Spacescited by 580
  19. 2024LLaVA-NeXT-Interleave: Tackling Multi-image, Video, and 3D in Large Multimodal Modelscited by 571
  20. 2024LLaVA-Video: Video Instruction Tuning With Synthetic Datacited by 550
  21. 2023VideoCrafter1: Open Diffusion Models for High-Quality Video Generationcited by 530
  22. 2023CoTracker: It is Better to Track Togethercited by 516
  23. 2024YOLO-World: Real-Time Open-Vocabulary Object Detectioncited by 498
  24. 2015A Neural Algorithm of Artistic Stylecited by 480
  25. 2022MagicVideo: Efficient Video Generation With Latent Diffusion Modelscited by 463
  26. 2023DynamiCrafter: Animating Open-domain Images with Video Diffusion Priorscited by 463
  27. 2023FateZero: Fusing Attentions for Zero-shot Text-based Video Editingcited by 462
  28. 2023TokenFlow: Consistent Diffusion Features for Consistent Video Editingcited by 440
  29. 2023LAVIE: High-Quality Video Generation with Cascaded Latent Diffusion Modelscited by 349
  30. 2023EvalCrafter: Benchmarking and Evaluating Large Video Generation Modelscited by 332
  31. 2023VideoFusion: Decomposed Diffusion Models for High-Quality Video Generationcited by 330
  32. 2023Pix2Video: Video Editing using Image Diffusioncited by 318
  33. 2023Video-P2P: Video Editing with Cross-attention Controlcited by 312
  34. 2022CogView2: Faster and Better Text-to-Image Generation via Hierarchical Transformerscited by 305
  35. 2023Show-1: Marrying Pixel and Latent Diffusion Models for Text-to-Video Generationcited by 300
  36. 2023Preserve Your Own Correlation: A Noise Prior for Video Diffusion Modelscited by 284
  37. 2023StableVideo: Text-driven Consistency-aware Diffusion Video Editingcited by 182
  38. 2024T2V-CompBench: A Comprehensive Benchmark for Compositional Text-to-video Generationcited by 177

50 most cited of 72 All 72 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 82; 2026: 207.

289 papers.

Sort:

By area:

  1. 2025Frame Context Packing and Drift Prevention in Next-Frame-Prediction Video Diffusion Modelscited by 126
  2. 2025Perception, Reason, Think, and Plan: A Survey on Large Multimodal Reasoning Modelscited by 89
  3. 2025OpenS2V-Nexus: A Detailed Benchmark and Million-Scale Dataset for Subject-to-Video Generationcited by 73
  4. 2025LongCat-Video Technical Reportcited by 73
  5. 2025WoW: Towards a World omniscient World model Through Embodied Interactioncited by 66
  6. 2025Radial Attention: O(nlog n) Sparse Attention with Energy Decay for Long Video Generationcited by 64
  7. 20253D and 4D World Modeling: A Surveycited by 64
  8. 2026Helios: Real Real-Time Long Video Generation Modelcited by 56
  9. 2025Captain Cinema: Towards Short Movie Generationcited by 48
  10. 2026Context Forcing: Consistent Autoregressive Video Generation with Long Contextcited by 47
  11. 2026World Action Models: The Next Frontier in Embodied AIcited by 46
  12. 2025Scaling Image and Video Generation via Test-Time Evolutionary Searchcited by 38
38 more
  1. 2026Infinite-World: Scaling Interactive World Models to 1000-Frame Horizons via Pose-Free Hierarchical Memorycited by 37
  2. 2026WBench: A Comprehensive Multi-turn Benchmark for Interactive Video World Model Evaluationcited by 34
  3. 2025VideoHallu: Evaluating and Mitigating Multi-modal Hallucinations on Synthetic Video Understandingcited by 32
  4. 2025VChain: Chain-of-Visual-Thought for Reasoning in Video Generationcited by 31
  5. 2026MIND: Benchmarking Memory Consistency and Action Control in World Modelscited by 31
  6. 2025PAI-Bench: A Comprehensive Benchmark For Physical AIcited by 30
  7. 2025Decoupled Distillation to Erase: A General Unlearning Method for Any Class-centric Taskscited by 28
  8. 2025FilMaster: Bridging Cinematic Principles and Generative AI for Automated Film Generationcited by 28
  9. 2025EchoMimicV3: 1.3B Parameters are All You Need for Unified Multi-Modal and Multi-Task Human Animationcited by 28
  10. 2025Uni-MMMU: A Massive Multi-discipline Multimodal Unified Benchmarkcited by 28
  11. 2026Rolling Sink: Bridging Limited-Horizon Training and Open-Ended Testing in Autoregressive Video Diffusioncited by 28
  12. 2025Pack and Force Your Memory: Long-form and Consistent Video Generationcited by 26
  13. 20253D Scene Generation: A Surveycited by 24
  14. 2025StreamAvatar: Streaming Diffusion Models for Real-Time Interactive Human Avatarscited by 24
  15. 2025VABench: A Comprehensive Benchmark for Audio-Video Generationcited by 22
  16. 2025WorldLens: Full-Spectrum Evaluations of Driving World Models in Real Worldcited by 22
  17. 2025NarrLV: Towards a Comprehensive Narrative-Centric Evaluation for Long Video Generationcited by 21
  18. 2025QVGen: Pushing the Limit of Quantized Video Generative Modelscited by 20
  19. 2025Macro-from-Micro Planning for High-Quality and Parallelized Autoregressive Long Video Generationcited by 19
  20. 2025T2AV-Compass: Towards Unified Evaluation for Text-to-Audio-Video Generationcited by 19
  21. 2025T2VWorldBench: A Benchmark for Evaluating World Knowledge in Text-to-Video Generationcited by 18
  22. 2026AVGen-Bench: A Task-Driven Benchmark for Multi-Granular Evaluation of Text-to-Audio-Video Generationcited by 18
  23. 2026Omni-WorldBench: Towards a Comprehensive Interaction-Centric Evaluation for World Modelscited by 17
  24. 2025DC-VideoGen: Efficient Video Generation with Deep Compression Video Autoencodercited by 15
  25. 2025Simulating the Visual World with Artificial Intelligence: A Roadmapcited by 15
  26. 2025Video Generation Models Are Good Latent Reward Modelscited by 15
  27. 2025Memorize-and-Generate: Towards Long-Term Consistency in Real-Time Video Generationcited by 15
  28. 2026RoboWM-Bench: A Benchmark for Evaluating World Models in Robotic Manipulationcited by 15
  29. 2026Vista4D: Video Reshooting with 4D Point Cloudscited by 15
  30. 2026MBench: A Comprehensive Benchmark on Memory Capability for Video World Modelscited by 15
  31. 2025PhysCorr: Dual-Reward DPO for Physics-Constrained Text-to-Video Generation with Automated Preference Selectioncited by 14
  32. 2025BLADE: Block-Sparse Attention Meets Step Distillation for Efficient Video Generationcited by 13
  33. 2025Attention Surgery: An Efficient Recipe to Linearize Your Video Diffusion Transformercited by 13
  34. 2025MoAlign: Motion-Centric Representation Alignment for Video Diffusion Modelscited by 13
  35. 2025BAgger: Backwards Aggregation for Mitigating Drift in Autoregressive Video Diffusion Modelscited by 13
  36. 2026HiAR: Efficient Autoregressive Long Video Generation via Hierarchical Denoisingcited by 13
  37. 2026Hallo-Live: Real-Time Streaming Joint Audio-Video Avatar Generation with Asynchronous Dual-Stream and Human-Centric Preference Distillationcited by 13
  38. 2026Alaya-EVOKE: From Linear-Scaling Supervision to Endless Worldcited by 13

50 most cited of 289 All 289 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2025Video-Bench: Human-Aligned Video Generation Benchmark
  2. 2025World Consistency Score: A Unified Metric for Video Generation Quality
  3. 2025V-ReasonBench: Toward Unified Reasoning Benchmark Suite for Video Generation Models
  4. 2026A Mechanistic View on Video Generation as World Models: State and Dynamics
  5. 2026PhyWorld: Physics-Faithful World Model for Video Generation
  6. 2025A Physical Coherence Benchmark for Evaluating Video Generation Models via Optical Flow-guided Frame Prediction
  7. 2025TiViBench: Benchmarking Think-in-Video Reasoning for Video Generative Models
  8. 2026From Evaluation to Enhancement: Benchmarking and Improving Think-with-Video Reasoning for Video Generative Models