PAIWorld: A 3D-Consistent World Foundation Model for Robotic Manipulation

• URL: http://arxiv.org/abs/2606.18375v3
• Date: Tue, 23 Jun 2026 16:50:09 GMT
• Title: PAIWorld: A 3D-Consistent World Foundation Model for Robotic Manipulation
• Abstract summary: We present PAIWorld, a framework that augments diffusion-transformer world models via three core components.<n>It achieves state-of-the-art multi-view 3D consistency on robotic benchmarks, ranking 1st on the WorldArena leaderboard and 2nd on the AgiBot-Challenge2026 leaderboard.
• Score: 51.385664546670284
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: World foundation models (WFMs) are powerful simulators, yet they predominantly operate in a single-view setting and lack the multi-view 3D consistency required for robotic manipulation. While robotic systems rely on multiple cameras (egocentric, eye-to-hand, and wrist-mounted) for policy learning, current multi-view world models simply concatenate view tokens without explicit geometric reasoning. This causes cross-view object drift, depth inconsistency, and texture misalignment. We trace these failures to two deficiencies: the absence of an explicit inter-view communication mechanism and the lack of a 3D geometric prior. We argue that resolving both simultaneously is necessary and sufficient. To address this, we present PAIWorld, a framework that augments diffusion-transformer world models via three core components: (1) Geometry-Aware Cross-View Attention blocks that establish an explicit pathway across views, (2) Geometric Rotary Position Embedding that encodes camera ray directions and extrinsic poses into the attention mechanism, and (3) Latent 3D-REPA, which distills 3D-aware features from frozen 3D foundation models to ensure 3D consistency. Built upon a DiT-based world foundation model, PAIWorld achieves state-of-the-art multi-view 3D consistency on robotic manipulation benchmarks, ranking 1st on the WorldArena leaderboard and 2nd on the AgiBot-Challenge2026 leaderboard, while enabling downstream applications such as model-based planning, world action models, and multi-view policy post-training.

Related papers

• LAWM-3D: Learning 3D-Aware Latent Actions from Human Videos for Generalizable Robot World Models [64.11117186468533]
  World models enable agents to perform forward rollout and planning without real-world interaction.<n>Most existing LAMs rely on single-view inputs and operate primarily in 2D pixel space.<n>We propose LAWM-3D, which introduces three tightly coupled key designs for learning 3D-aware latent actions.
  arXiv  Detail & Related papers  (2026-08-06T07:47:01Z)
• Axolotl3D: a Unified Framework for Faithful 3D Shape Completion [1.6934548577260198]
  We present Axolotl3D, a 3D generation model that conditions on images, visibility masks, camera parameters, and a partial point cloud.<n>A unified training strategy synthesizes diverse conditioning regimes from large-scale 3D data, enabling robust cross-modal reasoning.<n>Experiments on Toys4K and OmniObject3D demonstrate state-of-the-art performance under both clean and occluded settings.
  arXiv  Detail & Related papers  (2026-07-22T18:38:20Z)
• WorldAgents: Can Foundation Image Models be Agents for 3D World Models? [82.83725150353915]
  We demonstrate that 2D models do indeed encapsulate a grasp of 3D worlds.<n>By exploiting this understanding, our method successfully synthesizes expansive, realistic, and 3D-consistent worlds.
  arXiv  Detail & Related papers  (2026-03-20T07:22:41Z)
• OneWorld: Taming Scene Generation with 3D Unified Representation Autoencoder [90.8453349494245]
  We present OneWorld, a framework that performs diffusion directly within a coherent 3D representation space.<n>OneWorld generates high-quality 3D scenes with superior cross-view consistency compared to state-of-the-art 2D-based methods.
  arXiv  Detail & Related papers  (2026-03-17T03:43:37Z)
• PointWorld: Scaling 3D World Models for In-The-Wild Robotic Manipulation [48.807071017228964]
  We introduce PointWorld, a large pre-trained 3D world model that unifies state and action in a shared 3D space as 3D point flows.<n>With a real-time (0.1s) inference speed, PointWorld can be efficiently integrated in the model-predictive control (MPC) framework for manipulation.<n>We demonstrate that a single pre-trained checkpoint enables a real-world Franka robot to perform rigid-body pushing, deformable and articulated object manipulation.
  arXiv  Detail & Related papers  (2026-01-07T10:29:12Z)
• FantasyWorld: Geometry-Consistent World Modeling via Unified Video and 3D Prediction [13.098585993121722]
  We present FantasyWorld, a geometry-enhanced framework that augments frozen video foundation models with a trainable geometric branch.<n>Our approach introduces cross-branch supervision, where geometry cues guide video generation and video priors regularize 3D prediction.<n>Experiments show that FantasyWorld effectively bridges video imagination and 3D perception, outperforming recent geometry-consistent baselines in multi-view coherence and style consistency.
  arXiv  Detail & Related papers  (2025-09-25T22:24:23Z)
• Lift3D Foundation Policy: Lifting 2D Large-Scale Pretrained Models for Robust 3D Robotic Manipulation [30.744137117668643]
  Lift3D is a framework that enhances 2D foundation models with implicit and explicit 3D robotic representations to construct a robust 3D manipulation policy.<n>In experiments, Lift3D consistently outperforms previous state-of-the-art methods across several simulation benchmarks and real-world scenarios.
  arXiv  Detail & Related papers  (2024-11-27T18:59:52Z)
• PonderV2: Pave the Way for 3D Foundation Model with A Universal Pre-training Paradigm [111.16358607889609]
  We introduce a novel universal 3D pre-training framework designed to facilitate the acquisition of efficient 3D representation.<n>For the first time, PonderV2 achieves state-of-the-art performance on 11 indoor and outdoor benchmarks, implying its effectiveness.
  arXiv  Detail & Related papers  (2023-10-12T17:59:57Z)
• Unsupervised Learning of Visual 3D Keypoints for Control [104.92063943162896]
  Learning sensorimotor control policies from high-dimensional images crucially relies on the quality of the underlying visual representations. We propose a framework to learn such a 3D geometric structure directly from images in an end-to-end unsupervised manner. These discovered 3D keypoints tend to meaningfully capture robot joints as well as object movements in a consistent manner across both time and 3D space.
  arXiv  Detail & Related papers  (2021-06-14T17:59:59Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.