Subspace-Decomposed JEPAs: Disentangling Progression and Content in Latent World Models

• URL: http://arxiv.org/abs/2605.31111v1
• Date: Fri, 29 May 2026 10:23:48 GMT
• Title: Subspace-Decomposed JEPAs: Disentangling Progression and Content in Latent World Models
• Abstract summary: Joint-Embedding Predictive Architectures learn compact latent world models by predicting future embeddings.<n>We carve the JEPA latent into two subspaces with disjoint roles.<n>We prove that the two anti-collapse forces act on disjoint coordinates, so they compose additively rather than competing on the same dimensions.
• Score: 1.034052616244602
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Joint-Embedding Predictive Architectures (JEPAs) learn compact latent world models by predicting future embeddings, but no single coordinate of the latent is designated to encode task progression. We carve the JEPA latent into two orthogonal subspaces with disjoint roles: a low-dimensional progression subspace shaped by a cosine-margin triplet loss, and a high-dimensional content subspace regularised by the existing SIGReg objective of LeWM. We prove that the two anti-collapse forces act on disjoint coordinates, so they compose additively rather than competing on the same dimensions. Our method, SD-JEPA improves over the LeWM baseline on the majority of its control benchmarks at matched compute, and outperforms the strongest non-LeWM JEPA baseline on Push-T; a subspace-ablation falsifier confirms the split is the load-bearing ingredient. Beyond planning, the resulting 1-D angular progression coordinate functions as a scene-aware compass on the latent. It advances with task progress, regresses when the agent backtracks, and under controlled perturbations both spikes and relocalises to a semantically appropriate new task-phase sector, separating the moment of surprise from its meaning in a way that prediction-error scalars cannot. Three quantitative tests back this up: $|Δθ_t|$ outperforms the standard latent-prediction-error surprise at localising semantic events on 40 held-out cube episodes by up to +0.18 pooled AUROC (97.5% per-episode win rate at $\pm 1$-step tolerance); a within-episode linear probe across all four environments (40 episodes per env) shows the 8-dimensional progression subspace (4.2% of the latent) explains 72-95% of task-progress variance..

Related papers

• A*-Thought-V2: Efficient Latent Reasoning via Geometric Dynamics of LLM [40.07171947823021]
  Chain-of-Thought (CoT) improves the reasoning ability of Large Language Models (LLMs) but incurs substantial computation and context costs.<n>We present A*-Thought-V2, a geometric dynamics of LLM guided framework that models CoT as a hidden-state trajectory and replaces hard deletion with an explicit-implicit interleaved latent architecture.
  arXiv  Detail & Related papers  (2026-09-07T17:56:20Z)
• Cut-ViT: Task-Specific Model Pruning via Gram Anchoring Subspace Consistency [71.81785123423516]
  We propose a task-specific pruning pipeline, named Cut-ViT, for pruning visual foundation models.<n> Basis-agnostic and residual constraints are adopted to align the gram subspaces between the native and pruned DINOv3 models.<n>Experiments show that Cut-ViT requires approximately one minute on a single A100 GPU to obtainworks at various sparsity levels.
  arXiv  Detail & Related papers  (2026-08-28T11:25:36Z)
• GaussianDream++: Efficient 3D Gaussian World Modeling for Robotic Manipulation [55.47130289185285]
  Vision-Language-Action (VLA) policies have advanced language-conditioned robotic manipulation.<n>We present textbfMethodname, a compact, policy-native extension that inserts textbfWorld State Tokens and textbfWorld Prediction Tokens directly into the VLA backbone.
  arXiv  Detail & Related papers  (2026-08-26T11:41:55Z)
• Qantara: Bridge-Flow Training for Multi-Paradigm JEPA Control [5.357451930622806]
  Joint-Embedding Predictive Architectures underpin a growing family of latent world models for control from raw pixels.<n>A single checkpoint that serves both would defer this choice to inference, when deployment constraints determine which path wins.<n>We present Qantara, an end-to-end JEPA whose joint training objective pairs a Brownian-bridge interpolant between consecutive clean latents.
  arXiv  Detail & Related papers  (2026-07-06T12:12:50Z)
• UWM-JEPA: Predictive World Models That Imagine in Belief Space [0.2864713389096699]
  We introduce the Unitary World Model JEPA, a JEPA world model with a density-matrix latent on a joint system-environment space and a learned unitary predictor.<n>The construction preserves the joint-state spectrum exactly during rollout, so the predictor itself cannot dissipate the represented uncertainty.<n>For JEPA world models to imagine under partial observability, latent geometry and predictor dynamics matter, not frozen context-encoding capacity alone.
  arXiv  Detail & Related papers  (2026-05-25T00:28:51Z)
• High-Dimensional Latents Should Be Diagnosed Through Phase Structure [56.362776482614976]
  We study autoencoder and variational-autoencoder latent spaces through the lens of spin-glass theory.<n>We formalize a latent-space spin-glass dictionary for a fixed decoder.<n>We show that driving the latent system toward the edge-of-stability of the topological trivialization regime has concrete downstream consequences.
  arXiv  Detail & Related papers  (2026-05-23T06:17:23Z)
• iGSP:Implicit Gradient Subspace Projection for Efficient Continual Learning of Vision-Language Models [14.623213614166259]
  iGSP is a novel framework that achieves efficient adaptation via implicit gradient subspace projection.<n>Experiments on the MTIL benchmark demonstrate that iGSP achieves state-of-the-art accuracy.
  arXiv  Detail & Related papers  (2026-05-19T03:22:14Z)
• Looped SSMs: Depth-Recurrence and Input Reshaping for Time Series Classification [50.994194925685434]
  We show that a looped SSM with $k$ parameters iterated $L$ times consistently closely matches or outperforms a standard SSM.<n>We also show that input reshaping is an equally neglected design axis.
  arXiv  Detail & Related papers  (2026-05-15T15:18:12Z)
• In-Context Learning Operates as Concept Subspace Learning [20.051573945788963]
  We study whether structured demonstrations induce low-dimensional concept inference.<n>For ridge and least-squares ICL proxies, prediction decomposes exactly into concept-coordinate regression and off-subspace leakage.<n>These results support concept subspaces as compact, task-aligned mediators of recoverable ICL behavior.
  arXiv  Detail & Related papers  (2026-05-12T21:43:48Z)
• SpatialFly: Geometry-Guided Representation Alignment for UAV Vision-and-Language Navigation in Urban Environments [49.966170814478915]
  We propose a geometry-guided spatial representation framework for UAV VLN.<n>Running on RGB observations without explicit 3D reconstruction, SpatialFly introduces a geometry-guided 2D representation alignment mechanism.<n> Experimental results show that SpatialFly consistently outperforms state-of-the-art UAV VLN baselines across both seen and unseen environments.
  arXiv  Detail & Related papers  (2026-03-22T03:56:58Z)
• Probing the Latent World: Emergent Discrete Symbols and Physical Structure in Latent Representations [0.0]
  Video world models trained with Joint Embedding Predictive Architectures (JEPA) acquire rich representations by predicting masked regions in latent space rather than reconstructing pixels.<n>This pathway removes the visual verification of generative models, creating a structural interpretability gap: the encoder has learned physical structure inaccessible in any inspectable form.<n>We propose the AI Mother Tongue (AIM) framework as a passive quantization probe: a lightweight, vocabulary-free probe that converts V-JEPA 2 continuous latent vectors into discrete symbol sequences without task-specific supervision or modifying the encoder.
  arXiv  Detail & Related papers  (2026-03-20T01:30:59Z)
• LeWorldModel: Stable End-to-End Joint-Embedding Predictive Architecture from Pixels [49.35636088613484]
  Joint Embedding Predictive Architectures (JEPAs) offer a compelling framework for learning world models in compact latent spaces.<n>We introduce LeModelWorld, the first JEPA that trains stably end-to-end from raw pixels using only two loss terms.<n>With 15M parameters trainable on a single GPU in a few hours, LeWM plans up to 48x faster than foundation-model-based world models.
  arXiv  Detail & Related papers  (2026-03-13T19:48:14Z)
• Coordinate Transformer: Achieving Single-stage Multi-person Mesh Recovery from Videos [91.44553585470688]
  Multi-person 3D mesh recovery from videos is a critical first step towards automatic perception of group behavior in virtual reality, physical therapy and beyond. We propose the Coordinate transFormer (CoordFormer) that directly models multi-person spatial-temporal relations and simultaneously performs multi-mesh recovery in an end-to-end manner. Experiments on the 3DPW dataset demonstrate that CoordFormer significantly improves the state-of-the-art, outperforming the previously best results by 4.2%, 8.8% and 4.7% according to the MPJPE, PAMPJPE, and PVE metrics, respectively.
  arXiv  Detail & Related papers  (2023-08-20T18:23:07Z)
• Auxiliary Tasks Benefit 3D Skeleton-based Human Motion Prediction [106.06256351200068]
  This paper introduces a model learning framework with auxiliary tasks. In our auxiliary tasks, partial body joints' coordinates are corrupted by either masking or adding noise. We propose a novel auxiliary-adapted transformer, which can handle incomplete, corrupted motion data.
  arXiv  Detail & Related papers  (2023-08-17T12:26:11Z)
• Simple and Effective Prevention of Mode Collapse in Deep One-Class Classification [93.2334223970488]
  We propose two regularizers to prevent hypersphere collapse in deep SVDD. The first regularizer is based on injecting random noise via the standard cross-entropy loss. The second regularizer penalizes the minibatch variance when it becomes too small.
  arXiv  Detail & Related papers  (2020-01-24T03:44:47Z)

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.