arXiv:2609.38578v1 Announce Type: cross Abstract: Cross-structural motion retargeting aims to transfer motion between different skeletal topologies. Despite recent progress, existing state-of-the-art...
The paper examines latent action models that encode transitions between video frames using algebraic constraints such as additive composition and antisymmetric reversal. It demonstrates that these algebraic consistency conditions do not reliably certify temporal structure, as unconstrained models can achieve similar error reductions and constrained models still outperform unconstrained ones even after temporal pairings are destroyed. The authors find that preserving temporal pairing offers no consistent advantage on downstream tasks and that a direct repair objective yields only marginal improvement, recommending a more rigorous validation protocol.
The paper argues that diffusion-based action policies can use a frozen, observation‑free backbone as a reusable trajectory prior, with task adaptation handled entirely by the conditioning pathway. By pretraining a general action head on forward‑kinematics data and then freezing it, the authors show that a single backbone can match or outperform normally trained models on MimicGen and LIBERO. Their experiments reveal that a small 5 M‑parameter MLP backbone can rival large U‑Net and transformer backbones, indicating that action backbones are often over‑parameterized and that image‑style architectures may not be the best fit for low‑dimensional action generation.
arXiv:2607. 24083v1 Announce Type: new Abstract: Reinforcement learning can produce robust humanoid controllers, but each new task is typically trained as a separate policy with its own reward design and training process.
The paper introduces a unified conditional-flow framework that integrates text-driven motion generation, semantic editing, and intra-structural retargeting into a single rectified-flow model. By treating editing as a change in semantic condition and retargeting as a change in skeletal condition, the approach eliminates fragmented pipelines and allows a single model to perform generation, zero‑shot editing, and zero‑shot retargeting on articulated 3D motion data. Experiments on SnapMoGen and a Mixamo subset demonstrate that the model can handle all three tasks without task‑specific fine‑tuning, preserving both motion semantics and skeletal structure.
The paper introduces Action Forcing, a method that transforms ordinary unlabeled video into action‑supervised training data by extracting egomotion bases through principal component analysis of pixel displacements. This approach yields grounded throttle–yaw control signals without requiring instrumented platforms or manual annotation, and it trains a high‑capacity video model while preventing pixel‑level overfitting via an online latent critic. The authors also critique standard video generation metrics and propose a reference‑free evaluation that measures controllability, plausibility, conjuring, and geometric integrity, showing that their model can reverse, scale, and compose actions despite limited reverse‑action data.
The paper investigates 12‑class body‑only emotion recognition from skeleton motion using a leave‑performer‑out evaluation, where chance accuracy is 8.3% and a reproduced STGCN++ baseline scores 25.73% Macro‑F1. By ensembling eleven models with orthogonal error modes, the authors achieve 36.80% Macro‑F1, a 43% relative improvement over the baseline. They also introduce a tested explanation suite that demonstrates the ensemble’s decisions rely on motion‑grounded body‑region evidence, aligning strongly with Laban Movement Analysis attributes rather than classical kinematics, while showing diffuse temporal saliency.
LPA-CWM introduces a Learned Physical Adjudicator (LPA) to improve counterfactual world models (CWM) for motion reasoning by learning to weight candidate responses based on visual context and response structure. The 3.0M‑parameter LPA is trained on dense MOVi‑F trajectories while keeping the CWM predictor and intervention generator frozen. A new Completeness‑aware Motion Correspondence (CMC) protocol evaluates localization, trajectory completeness, visibility, and continuity, and LPA‑CWM achieves significant gains on DAVIS and Kinetics subsets.
arXiv:2604. 21241v2 Announce Type: replace-cross Abstract: Vision--Language--Action (VLA) models often use intermediate representations to connect multimodal inputs with continuous control, yet spatial guidance is often injected implicitly through latent features.
The paper introduces Action-Contrastive Masked Transition Modeling (AC‑MTM), a method that stabilizes Joint‑Embedding Predictive Architectures (JEPAs) without relying on Gaussian regularization. AC‑MTM adds a training‑only inverse‑dynamics head that uses Action‑NCE to force each latent transition to identify its generating action, thereby preventing encoder collapse. Experiments on pixel‑control and multi‑object visual tasks show that AC‑MTM trains stably from scratch and matches or surpasses the performance of SIGReg, achieving up to a 24‑point improvement on the OGBench Visual Scene benchmark.
arXiv:2607. 10206v1 Announce Type: cross Abstract: Flow-matching policies are promising for imitation learning because they model complex multimodal action distributions.
ActionPiece rethinks how actions are tokenized for autoregressive vision‑language‑action models by introducing physical rank consistency (PRC) to evaluate relational fidelity of reconstructed actions. The method jointly supervises representation learning and quantization to preserve local physical distance rankings, improving both PRC and policy success. Experiments on LIBERO, LIBERO‑Plus, SimplerEnv, and VLA‑Arena show significant gains over baseline tokenizers.