AcrossWAM1.0:A Modular Latent World-Action Stack for Compact Robot Policies

• URL: http://arxiv.org/abs/2608.29937v1
• Date: Sun, 30 Aug 2026 18:09:17 GMT
• Title: AcrossWAM1.0:A Modular Latent World-Action Stack for Compact Robot Policies
• Abstract summary: We introduce AcrossWAM1.0, a modularization and scaling study of LaWAM's latent world-action stack.<n>AcrossWAM1.0 contributes an auditable software and evaluation boundary for compact latent world-action policies.
• Score: 14.556446786016787
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Latent world-action models avoid rendering future pixels by predicting an action-relevant visual subgoal in feature space. LaWAM established this formulation, but its original presentation left the world model, multimodal backbone, and deployment checkpoint tightly coupled. We introduce AcrossWAM1.0, a modularization and scaling study of this latent world-action stack. Rather than presenting latent subgoals as a new algorithm, we make the module boundary explicit: a policy adapter produces latent-action and action-generation contexts; a retained latent world decoder grounds the predicted transition in the current scene;and a flow-matching expert generates continuous action chunks. We further separate training-only teachers from the inference graph and provide a verifiable deployment export. On 2,000 paired LIBERO episodes, replacing a Qwen3-VL-2B backbone with Qwen3.5-0.8B yields 97.45% success versus 98.00% for the 2B model (a-0.55percentage-point difference; exact McNemarp=0.266). This does not prove equivalence, but it meets a prespecified two-point retention criterion. The compact, inference-reachable checkpoint contains 1,472.6M unique parameters, 42.4% fewer than the original 2B policy, while all retained tensors are bitwise identical to the source checkpoint. Cross-family execution is additionally checked with a MiniCPM-V adapter smoke test; closed-loop cross-family transfer remains an open evaluation. AcrossWAM1.0 therefore contributes an auditable software and evaluation boundary for compact latent world-action policies, distinct from LaWAM's original latent-subgoal contribution.

Related papers

• ReDraft, Don't Just Distill: Reference-Driven Revision for Continual VLLM Post-Training [66.72708893922605]
  Post-training of large multimodal models should add new capabilities while preserving those from pre-training.<n>SFT gives explicit target supervision that learns a task from near-zero accuracy, but its off-policy targets move the model far enough to cause forgetting.<n>We introduce ReDraft, which obtains both from the model's own failures.
  arXiv  Detail & Related papers  (2026-09-15T04:59:03Z)
• Schema-Adaptive Action-Conditioned JEPA for Cross-Machine CNC Transfer under Partial Sensor Overlap [0.0]
  Cross-machine deployment of industrial world models requires transfer across changes in dynamics, sensing interfaces, sampling regimes, and control units.<n>We study a schema-adaptive action-conditioned Joint-Embedding Predictive Architecture (SAAC-JEPA) for CNC dynamics.<n> Evaluation uses group-disjoint source splits, source-only normalization, held-out self-supervised validation, unit audits, and a sealed target test after model locking.
  arXiv  Detail & Related papers  (2026-09-13T15:10:06Z)
• Logit Refiner: Improving Visual Autoregressive Models via Intra-Scale Dependency Modeling [21.906953897227442]
  Visual Autoregressive Models generate images through next-scale prediction.<n>Mean-field approximation discards spatial dependencies among same-scale tokens.<n>Logit Refiner restores intra-scale dependencies by sequentially sampling tokens conditioned on frozen backbone features.
  arXiv  Detail & Related papers  (2026-09-10T16:57:05Z)
• Selective Cross-View Consistency for World Action Models: Held-Out Viewpoint Robustness Without Test-Time Camera Information [0.0]
  World action models (WAMs) jointly denoise future video frames and robot actions, and the video prior is expected to generalize their control.<n>We study a question specific to this model class: when training with same-state cross-view image pairs, on which output coordinates should a consistency loss be imposed?<n>We show that consistency applied to the covariant block is provably harmful, shrinking legitimate view-specific content to a fraction $1/(1+4)$ of its true value, and we verify this shrinkage law in controlled experiments.
  arXiv  Detail & Related papers  (2026-08-07T12:03:23Z)
• QuantWAMs: Calibrating at the Right Granularity for World Action Models [34.540026172139406]
  World Action Models jointly predict future observations and actions, but their iterative denoising and closed-loop execution make deployment costly.<n>We present QuantWAMs, a PTQ framework that aligns quantization decisions with the calibration context defined by model structure, rollout distribution, and task objective.<n>We evaluate QuantWAMs on Fast-WAM and LingBot-VA across RoboTwin 2.0, LIBERO, and real-robot manipulation with an AgiBot G2.
  arXiv  Detail & Related papers  (2026-07-30T15:54:29Z)
• Metric Aggregation Divergence: A Hidden Validity Threat in Agent-Based Policy Optimization and a Contractual Remedy [7.1976264551575895]
  Metric aggregation divergence (MAD) is the silent inconsistency that arises when distinct pipeline stages in an agent-based model.<n>We introduce the metric contract - a single callable runtime enforced at dispatch time - as the remedy.
  arXiv  Detail & Related papers  (2026-06-27T18:17:33Z)
• Label-Free Cross-Task LoRA Merging with Null-Space Compression [50.63908869296697]
  We introduce Null-Space Compression (NSC) Merging, a label-free, output-agnostic method that sets merge weights from adapter geometry.<n>NSC achieves state-of-the-art performance across twenty heterogeneous vision tasks with balanced gains where prior methods overfit subsets of tasks.
  arXiv  Detail & Related papers  (2026-03-27T11:34:41Z)
• See Less, Drive Better: Generalizable End-to-End Autonomous Driving via Foundation Models Stochastic Patch Selection [51.59559387222532]
  Recent advances in end-to-end autonomous driving show that policies trained on patch-aligned features generalize better to Out-of-Distribution (OOD)<n>We present $2.4-Patch-Selection (SPS), a simple yet effective approach for learning policies that are more robust, generalizable, and efficient.
  arXiv  Detail & Related papers  (2026-01-15T18:58:33Z)
• Variational Gaussian Mixture Manifold Models for Client-Specific Federated Personalization [0.0]
  VGM$2$ is a geometry-centric PFL framework that learns client-specific parametric UMAP embeddings.<n>Each client maintains a Dirichlet-Normal-Inverse-Gamma posterior over marker weights, means, and variances.<n>VGM$2$ achieves competitive or superior test F1 scores compared to strong baselines.
  arXiv  Detail & Related papers  (2025-09-04T01:28:02Z)
• MASS: MoErging through Adaptive Subspace Selection [55.03293736484465]
  We present MASS (MoErging through Adaptive Subspace Selection), a new approach to model merging.<n> MASS stores only the most salient singular components for each task and merges them into a shared model.<n>We evaluate MASS on CLIP-based image classification using ViT-B-16, ViT-B-32 and ViT-L-14 for benchmarks of 8, 14 and 20 tasks respectively.
  arXiv  Detail & Related papers  (2025-04-06T08:49:52Z)
• PointOBB-v3: Expanding Performance Boundaries of Single Point-Supervised Oriented Object Detection [65.84604846389624]
  We propose PointOBB-v3, a stronger single point-supervised OOD framework.<n>It generates pseudo rotated boxes without additional priors and incorporates support for the end-to-end paradigm.<n>Our method achieves an average improvement in accuracy of 3.56% in comparison to previous state-of-the-art methods.
  arXiv  Detail & Related papers  (2025-01-23T18:18:15Z)
• PointPatchMix: Point Cloud Mixing with Patch Scoring [58.58535918705736]
  We propose PointPatchMix, which mixes point clouds at the patch level and generates content-based targets for mixed point clouds. Our approach preserves local features at the patch level, while the patch scoring module assigns targets based on the content-based significance score from a pre-trained teacher model. With Point-MAE as our baseline, our model surpasses previous methods by a significant margin, achieving 86.3% accuracy on ScanObjectNN and 94.1% accuracy on ModelNet40.
  arXiv  Detail & Related papers  (2023-03-12T14:49:42Z)
• A Coarse-to-Fine Instance Segmentation Network with Learning Boundary Representation [10.967299485260163]
  Boundary-based instance segmentation has drawn much attention since of its attractive efficiency. Existing methods suffer from the difficulty in long-distance regression. We propose a coarse-to-fine module to address the problem.
  arXiv  Detail & Related papers  (2021-06-18T16:37:28Z)

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.