Spatial Lifting (SL) is a new method for dense prediction that lifts 2D inputs into a higher-dimensional space and processes them with networks suited to that dimension, such as a 3D U‑Net. This dimensionality lifting surprisingly yields competitive performance while cutting inference costs and dramatically reducing model parameters. The SL framework naturally produces structured outputs along the lifted dimension, enabling dense supervision during training and single‑forward‑pass self‑consistency‑based quality and uncertainty estimation at test time.
arXiv:2610.34473v1 Announce Type: new Abstract: Significant advancements in 3D asset creation have enabled high-fidelity 3D generation from images. However, most existing representations remain inher...
The paper introduces Spatially Aware World Action Model (SA‑WAM), a diffusion‑based framework that extends existing World Action Models by incorporating depth information alongside RGB to enable 3‑D‑aware action and future‑state prediction. SA‑WAM repurposes a pretrained video diffusion model, using a nonlinear encoding to map unbounded depth into the tokenizer’s bounded domain, thus preserving pretrained visual priors without 3‑D‑specific fine‑tuning. The model achieves state‑of‑the‑art performance on RoboCasa and LIBERO‑Plus benchmarks and demonstrates superior real‑world performance on a UR5 robotic arm in randomized environments, while also providing analysis linking world‑model prediction quality to rollout success.
Recent advances in diffusion models have shown impressive performance in controllable image generation and dense prediction tasks. However, existing approaches typically treat diffusion-based controllable generation and dense prediction as separate tasks, overlooking the potential benefits of jointly modeling the heterogeneous distributions.
arXiv:2608.14740v2 Announce Type: replace Abstract: Unified image and video creation requires a model to follow diverse instructions while preserving identity, geometry, and temporal structure from v...
The paper introduces GeoNeXt, a framework that repurposes pretrained video generative models for geometry estimation by framing it as a next‑frame prediction task. Unlike prior methods that either train separate depth/normal models or fine‑tune image diffusion backbones, GeoNeXt jointly models images and geometric targets, leveraging the structured knowledge of video models for more data‑efficient learning. Experiments show zero‑shot monocular depth and surface normal estimation that outperforms existing generative approaches and rivals discriminative state‑of‑the‑art methods while using far less training data.
The paper introduces MaSS, a framework that adds a dedicated structural representation pathway to self‑supervised monocular depth estimation. By using a scene‑structure discrimination objective, MaSS learns generalized, layout‑aware structural representations that decouple from contextual features. Experiments on four diverse datasets show MaSS outperforms prior methods and benefits from mixed‑domain training, while zero‑shot tests on eight additional datasets confirm its strong generalizability to out‑of‑distribution scenes.
DeepStratNet introduces a context‑aware coordinate regression framework for seismic horizon tracking that directly predicts time/depth coordinates at each lateral position, avoiding the need for dense segmentation masks. The lightweight regression head, combined with an LSTM for inter‑slice context and geology‑informed regularization, outperforms traditional segmentation models on a New Zealand seismic volume, especially under sparse labeling. The method also provides a built‑in quality control by capturing local geological variations through prediction variability across traces.
arXiv:2610.16817v1 Announce Type: cross Abstract: Progressive depth growth builds large networks by starting with shallow models and inserting layers during training. Recent advances in large languag...
arXiv:2603. 19312v3 Announce Type: replace Abstract: Joint Embedding Predictive Architectures (JEPAs) offer a compelling framework for learning world models in compact latent spaces, yet existing methods remain fragile, relying on complex multi-term losses, exponential moving averages, pre-trained encoders, or auxiliary supervision to avoid representation collapse.
arXiv:2602. 14134v2 Announce Type: replace-cross Abstract: Multimodal Large Language Models (MLLMs) have demonstrated exceptional capabilities in high-level visual understanding.
arXiv:2607. 09892v1 Announce Type: cross Abstract: We introduce DenseAR, a new generative paradigm that reformulates autoregressive image generation as coarse-to-fine next-dense-stride prediction using a compact single-scale tokenizer.
arXiv:2606. 29579v1 Announce Type: cross Abstract: Spatial reasoning remains a persistent challenge for many vision language models (VLMs), and improving it typically requires fine-tuning with substantial additional parameters.