Research map / Vision-language and multimodal / Multimodal LLMs
Simulating the Real World: A Unified Survey of Multimodal Generative Models
Yuqi Hu, Longguang Wang, Xian Liu, Ling-Hao Chen, Yuwei Guo, Yukai Shi, Ce Liu, Anyi Rao, Zeyu Wang, Hui Xiong
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use a feedforward generator (Trellis) instead of optimization-based ProlificDreamer | text-to-3D · T3Bench (Single Object) | time: ~240 min -> <1 min | Improved |
| use a feedforward generator (Trellis) instead of optimization-based ProlificDreamer | text-to-3D · T3Bench (Single Object) | quality: 51.1 -> 35.6 | Hurt |
| use the multi-view-based MVDream instead of DreamFusion | text-to-3D · T3Bench (Single Object) | quality: 24.9 -> 53.2 | Improved |
| use DreamFusion instead of Fantasia3D | text-to-3D · T3Bench (Single Object) | alignment: 23.5 -> 24.0 alignment, 29.2 -> 24.9 quality | Mixed |
| use InstantMesh (MVS-based) instead of Realfusion (optimization-based) | image-to-3D · GSO | Chamfer distance: 0.0819 -> 0.0187 | Improved |
| use InstantMesh instead of Realfusion | image-to-3D · GSO | IoU: 0.2741 -> 0.6353 | Improved |
| use feedforward CraftsMan3D instead of Point-E | image-to-3D · GSO | Chamfer distance: 0.0426 -> 0.0291 | Improved |
| use STAG4D instead of Consistent4D | 4D generation · 4D generation | CLIP: 0.87 -> 0.91 | Improved |
| use STAG4D instead of Consistent4D | 4D generation · 4D generation | LPIPS: 0.16 -> 0.13 | Improved |
| use DG4D instead of Consistent4D | 4D generation · 4D generation | time: 2 hr -> 10 min | Improved |
| use Sora instead of Mochi | text-to-video · text-to-video benchmark | Motion Quality: 0.420 -> 0.482 | Improved |
| use Sora instead of Mochi | text-to-video · text-to-video benchmark | Pixel Stability: 0.981 -> 0.952 | Hurt |
Closest papers by meaning
- 4DWorldBench: A Comprehensive Evaluation Framework for 3D/4D World Generation Models (CVPR 2026 · similarity 0.91)
- Learning Interactive Real-World Simulators (ICLR 2024 · similarity 0.89)
- DuoGen: Towards Autonomous Interleaved Multimodal Generation (CVPR 2026 · similarity 0.88)
- Learning 3D Persistent Embodied World Models (NeurIPS 2025 · similarity 0.88)
- DynamicVerse: A Physically-Aware Multimodal Framework for 4D World Modeling (NeurIPS 2025 · similarity 0.88)
- GGBench: A Geometric Generative Reasoning Benchmark for Unified Multimodal Models (CVPR 2026 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
