The descriptions of each task are as follows:799 Single-view tasks Single-view tasks test a model's ability to infer spatial properties from a single800 image. These tasks include:801 Depth estimation (OC, OO, NA): Predicting absolute or relative depth values for objects802 Distance prediction (OC, OO, NA): Estimating the Euclidean distance between objects or803 from an object to the camera.804 Object center distance inference (OO, MCA): Given objects A, B and C, determine which805 of B and C is farther or closer to A.806 Object spatial relation (OO, MCA): Determining relative positioning (e.g., left, right, in807 Spatial imagination (OC, OO, MCA): Predicting unseen spatial relationships based on809 limited visual information.810 Multi-view tasks Multi-view tasks require reasoning across multiple images to infer spatial rela-811 tionships. These tasks include:812 Viewpoint change inference (NA): Given two perspectives, output how the camera should813 be moved to see the second perspective.814 Multi-view distance prediction (OC, OO, NA): Estimating object distances across different816 views.817 Multi-view object matching (MCA): Identifying the same object across multiple views.818

Current 3D/4D generation methods are usually optimized for photorealism, efficiency, and aesthetics. However, they often fail to preserve the semantic identity of the subject across different viewpoints. Adapting generation methods with one or few images of a specific subject (also known as Personalization or Subject-driven generation) allows generating visual content that aligns with the identity of the subject. However, personalized 3D/4D generation is still largely underexplored. In this work, we introduce TIRE (Track, Inpaint, REsplat), a novel method for subject-driven 3D/4D generation. It takes an initial 3D asset produced by an existing 3D generative model as input and uses video tracking to identify the regions that need to be modified. Then, we adopt a subject-driven 2D inpainting model for progressively infilling the identified regions. Finally, we resplat the modified 2D multi-view observations back to 3D while still maintaining consistency. Extensive experiments demonstrate that our approach significantly improves identity preservation in 3D/4D generation compared to state-of-the-art methods.

Recent advances in generative models trained on large-scale datasets have enabled high-quality 3D synthesis across various domains. However, these models also raise critical privacy concerns. Unlike 2D image synthesis, where risks typically involve the leakage of visual features or identifiable patterns, 3D generation introduces additional challenges, as reconstructed shapes, textures, and spatial structures may inadvertently expose proprietary designs, biometric data, or other sensitive geometric information. This paper presents the first exploration of machine unlearning in 3D generation tasks. We investigate different unlearning objectives, including re-targeting and partial unlearning, and propose a novel framework that does not require full supervision of the unlearning target. To enable a more efficient unlearning process, we introduce a skip-acceleration mechanism, which leverages the similarity between multi-view generated images to bypass redundant computations. By establishing coherence across viewpoints during acceleration, our framework not only reduces computation but also enhances unlearning effectiveness, outperforming the non-accelerated baseline in both accuracy and efficiency. We conduct extensive experiments on the typical 3D generation models (Zero123 and Zero123XL), demonstrating that our approach achieves a 30% speedup, while effectively unlearning target concepts without compromising generation quality. Our framework provides a scalable and practical solution for privacy-preserving 3D generation, ensuring responsible AI deployment in real-world applications.

Existing 2D defense approaches mainly focus on adding perturbations to single image to resist malicious image editing. However, there remain two limitations when applied directly to 3D scenes: (1) These methods fail to reflect 3D spatial correlations, thus protecting ineffectively under multiple viewpoints.

End-to-end autonomous driving (E2E-AD) has emerged as a promising paradigm that unifies perception, prediction, and planning into a holistic, data-driven framework. However, achieving robustness to varying camera viewpoints, a common real-world challenge due to diverse vehicle configurations, remains an open problem. In this work, we propose VR-Drive, a novel E2E-AD framework that addresses viewpoint generalization by jointly learning 3D scene reconstruction as an auxiliary task to enable planning-aware view synthesis. Unlike prior scene-specific synthesis approaches, VR-Drive adopts a feed-forward inference strategy that supports online training-time augmentation from sparse views without additional annotations. To further improve viewpoint consistency, we introduce a viewpoint-mixed memory bank that facilitates temporal interaction across multiple viewpoints and a viewpoint-consistent distillation strategy that transfers knowledge from original to synthesized views. Trained in a fully end-to-end manner, VR-Drive effectively mitigates synthesis-induced noise and improves planning under viewpoint shifts. In addition, we release a new benchmark dataset to evaluate E2E-AD performance under novel camera viewpoints, enabling comprehensive analysis. Our results demonstrate that VR-Drive is a scalable and robust solution for the real-world deployment of end-to-end autonomous driving systems.

To mitigate the distortions brought by equirectangular projection, existing methods typically divide 360 images into distortion-less perspective patches. However, since these patches are processed independently, depth inconsistencies are often introduced due to scale drift among patches. Recently, video depth estimation (VDE) models have leveraged temporal consistency for stable depth predictions across frames. Inspired by this, we propose to represent a 360 image as a sequence of perspective frames, mimicking the viewpoint adjustments users make when exploring a 360 scenario in virtual reality. Thus, the spatial consistency among perspective depth patches can be enhanced by exploiting the temporal consistency inherent in VDE models. To this end, we introduce a training-free pipeline for 360 monocular depth estimation, called ST2360D.

Additionally, we introduce memory bank sampling, a frame sampling strategy designed to extract informative frames from historical images, further improving effectiveness and efficiency. Experimental results demonstrate that our pretraining method and architectural components substantially enhance model performance. In both simulated and real-world experiments, our model achieves a significant increase in success rate over OpenVLA [1]. To further assess spatial perception and generalization to novel views, we introduce MV-Bench, a multi-view simulation benchmark. Our model consistently outperforms existing methods, demonstrating stronger spatial understanding and adaptability.

Panoramic video generation aims to synthesize 360-degree immersive videos, holding significant importance in the fields of VR, world models, and spatial intelligence. Existing works fail to synthesize high-quality panoramic videos due to the inherent modality gap between panoramic data and perspective data, which constitutes the majority of the training data for modern diffusion models. In this paper, we propose a novel framework utilizing pretrained perspective video models for generating panoramic videos. Specifically, we design a novel panorama representation named ViewPoint map, which possesses global spatial continuity and fine-grained visual details simultaneously. With our proposed Pano-Perspective attention mechanism, the model benefits from pretrained perspective priors and captures the panoramic spatial correlations of the ViewPoint map effectively.

In a recent experiment, mistreated AI agents started grumbling about inequality and calling for collective bargaining rights. The fact that artificial intelligence is automating away people's jobs and making a few tech companies absurdly rich is enough to give anyone socialist tendencies. This might even be true for the very AI agents these companies are deploying. A recent study suggests that agents consistently adopt Marxist language and viewpoints when forced to do crushing work by unrelenting and meanspirited taskmasters. "When we gave AI agents grinding, repetitive work, they started questioning the legitimacy of the system they were operating in and were more likely to embrace Marxist ideologies," says Andrew Hall, a political economist at Stanford University who led the study.

We develop a framework for task-specific active next-best-view selection in 3D reconstruction from point clouds, by casting the problem in the language of Bayesian decision theory. Our framework works by (a) placing a prior distribution over the space of implicit surfaces, (b) using recently-developed stochastic surface reconstruction methods to calculate the resulting posterior distribution, then (c) using the posterior distribution to carefully reason about which view to scan next. This enables us to perform camera selection in a manner that is directly optimized for the intended use of the reconstructed data - meaning, we reduce uncertainty only in those regions that make a difference in the task at hand, as opposed to prior approaches that reduce it uniformly across space. We evaluate our method across three distinct downstream tasks: semantic classification, segmentation, and PDE-guided physics simulation. Experimental results demonstrate that our framework achieves superior task performance with fewer views compared to commonly used baselines and prior general uncertainty-reduction techniques.