Dream.exe evaluates 8 video generation models on 101 manipulation tasks by converting generated videos into executable robot trajectories in a simulator, finding measurable success rates that visual metrics do not predict.
arXiv preprint arXiv:2602.08025 (2026)
9 Pith papers cite this work. Polarity classification is still indexing.
9
Pith papers citing it
fields
cs.CV 9years
2026 9verdicts
UNVERDICTED 9representative citing papers
WBench is a benchmark with 289 test cases and 1,058 turns for evaluating interactive world models using 22 automated metrics validated against human judgments.
Formalizes video world models as group actions on states and uses latent regularization with synthesized supervision to enforce consistency, introducing GAC and GAR metrics that improve structural correctness in SOTA models.
WorldMark is the first public benchmark that standardizes scenes, trajectories, and control interfaces across heterogeneous interactive image-to-video world models.
WorldOdysseyBench introduces four new evaluation dimensions and metrics for interactive world models and shows that none of 10+ tested models reliably pass all of them.
Current world models fail to evolve internal state when unobserved and instead resume scenes at the last observed state, as diagnosed by the new WRBench benchmark across 23 models and 9600 videos.
GIM-World adds a camera-queryable geometry distillation head and pruning rule to implicit memory in video world models, claiming better long-horizon geometric consistency on the MIND benchmark than explicit and implicit baselines.
PhysRAG curates 7K videos from WISA-80K, builds a physical video database, and injects knowledge via learnable queries into a diffusion model to reach SOTA visual quality and physical compliance on PhyGenBench and VBench.
WorldOlympiad is a new benchmark decomposing world-model evaluation into physical, geometry, and interaction tracks using segmentation, MLLM judges, Gaussian splatting, and action prompts on diverse scenarios.
citing papers explorer
-
Dream.exe: Can Video Generation Models Dream Executable Robot Manipulation?
Dream.exe evaluates 8 video generation models on 101 manipulation tasks by converting generated videos into executable robot trajectories in a simulator, finding measurable success rates that visual metrics do not predict.
-
WBench: A Comprehensive Multi-turn Benchmark for Interactive Video World Model Evaluation
WBench is a benchmark with 289 test cases and 1,058 turns for evaluating interactive world models using 22 automated metrics validated against human judgments.
-
World Models as Group Actions
Formalizes video world models as group actions on states and uses latent regularization with synthesized supervision to enforce consistency, introducing GAC and GAR metrics that improve structural correctness in SOTA models.
-
WorldMark: A Unified Benchmark Suite for Interactive Video World Models
WorldMark is the first public benchmark that standardizes scenes, trajectories, and control interfaces across heterogeneous interactive image-to-video world models.
-
WorldOdysseyBench: An Open-World Benchmark for Long-Horizon Stability of Interactive World Models
WorldOdysseyBench introduces four new evaluation dimensions and metrics for interactive world models and shows that none of 10+ tested models reliably pass all of them.
-
Current World Models Lack a Persistent State Core
Current world models fail to evolve internal state when unobserved and instead resume scenes at the last observed state, as diagnosed by the new WRBench benchmark across 23 models and 9600 videos.
-
Geometry-Aware Implicit Memory for Video World Models
GIM-World adds a camera-queryable geometry distillation head and pruning rule to implicit memory in video world models, claiming better long-horizon geometric consistency on the MIND benchmark than explicit and implicit baselines.
-
PhysRAG: Enhancing Physics-Awareness in Video Generation via Retrieval-Augmented Generation
PhysRAG curates 7K videos from WISA-80K, builds a physical video database, and injects knowledge via learnable queries into a diffusion model to reach SOTA visual quality and physical compliance on PhyGenBench and VBench.
-
WorldOlympiad: Can Your World Model Survive a Triathlon?
WorldOlympiad is a new benchmark decomposing world-model evaluation into physical, geometry, and interaction tracks using segmentation, MLLM judges, Gaussian splatting, and action prompts on diverse scenarios.