ScIntBench is a new benchmark for evaluating text‑conditioned 3D indoor scene generation, focusing on interaction readiness rather than just static physical plausibility. It runs prescribed manipulation actions—such as pushing, opening, and probing—in a physics simulation and records metrics like task success, capability, progress, post‑interaction integrity, and dynamic plausibility, supplemented by multi‑view semantic and static physical checks. Across 12 state‑of‑the‑art systems and 10 room types, the study finds that high static plausibility does not guarantee interaction readiness, with the best system succeeding in only 36.92% of tasks and many systems unable to support any interaction at all.
arXiv:2602. 09153v2 Announce Type: replace-cross Abstract: Simulation has become a key tool for training and evaluating home robots at scale, yet existing environments fail to capture the diversity and physical complexity of real indoor spaces.
arXiv:2605.15843v2 Announce Type: replace Abstract: Recent 3D world modeling systems based on generative scene synthesis, such as Marble, can create coherent and explorable 3D environments, yet their...
arXiv:2606. 31672v2 Announce Type: replace-cross Abstract: Despite rapid progress in interactive world models (IWMs), existing benchmarks evaluate action following only at trajectory level and ignore memory and interaction physics.
LLM agents are increasingly used to translate natural language into 3D scenes in a procedural way, but existing systems focus on static output. Dynamic 4D scenes from text alone, in which liquids flow, particles emit, rigid bodies cascade, and articulated mechanisms move, remain largely unexplored despite their value as editable content and as physics-grounded training data for video generation and embodied AI.
Despite rapid progress in interactive world models (IWMs), existing benchmarks evaluate action following only at trajectory level and ignore memory and interaction physics. We introduce WorldOdysseyBench, an open-world benchmark for long-horizon stability across four dimensions, each with tailored innovations: (i) Action: per-frame action metric bypassing cross-model semantic scale disparity and exposing failures hidden by trajectory; (ii) Vision: segment-based drift metric capturing non-monotonic mid-sequence collapse missed by start-vs-end comparisons; (iii) Physics: controllability-gated evaluation over mechanics, optics, and 3D consistency, scoring plausibility under faithful action execution; (iv) Memory: action-decoupled protocol evaluating scene memory via transition-localized 3D point-cloud reconstruction and subject memory via tracking-plus-VLM reasoning.
SceneTeract is a verification interface that separates semantic action understanding from physical feasibility in indoor 3D scenes. It decomposes activities into atomic actions and performs explicit geometric checks to determine executability, providing diagnostic traces for failures. The system reveals widespread functional and accessibility issues in synthetic scenes, shows that existing VLMs over‑predict action feasibility, and improves VLM performance through post‑training with verifier feedback, with benefits that generalize to real‑world scenes.
arXiv:2607. 01766v1 Announce Type: new Abstract: LLM agents are increasingly used to translate natural language into 3D scenes in a procedural way, but existing systems focus on static output.
arXiv:2610.15228v1 Announce Type: new Abstract: Words are increasingly becoming an interface for action: large language models (LLMs) can now invoke tools, operate devices, and make decisions that af...
WorldOracle is a benchmark that evaluates action‑conditioned world models (ACWMs) by comparing their predicted rollouts to paired real‑world executions. It assesses models across four dimensions—physical state correctness, spatial alignment, key interaction fidelity, and multi‑view coherence—using curated training and evaluation suites that include long, multi‑segment episodes with both successful and failed actions. Experiments on seven ACWMs highlight distinct strengths and failure modes, providing a systematic foundation for advancing manipulation modeling.
MAMHOI introduces an affordance-mediated factorization for generating scene-aware human‑object interactions. It separates scene understanding and motion synthesis: a scene‑conditioned model predicts feasible interaction locations and modes, while an affordance‑conditioned model generates the corresponding human‑object motion. This approach enables learning from separate human‑scene and human‑object datasets, reducing object‑scene penetration and improving interaction realism in complex indoor environments.
arXiv:2606. 31672v1 Announce Type: cross Abstract: Despite rapid progress in interactive world models (IWMs), existing benchmarks evaluate action following only at trajectory level and ignore memory and interaction physics.
World2HOSI is a multi‑expert harness that generates long‑horizon human‑object‑scene interaction (HOSI) data by orchestrating existing human‑object interaction (HOI) and human‑scene interaction (HSI) models. It builds a persistent scene state from decoupled background and object assets, plans action stages, invokes experts via their native interfaces, and updates the scene after each action, ensuring physical plausibility through bounded residual optimization and contact‑driven physics simulation. Experiments and a perceptual user study demonstrate that World2HOSI achieves superior task completion and interaction quality while producing more diverse data than prior approaches.