UniLACT improves VLA models by adding depth-aware unified latent action pretraining that outperforms RGB-only baselines on seen and unseen manipulation tasks.
arXiv preprint arXiv:2505.08787 , year=
9 Pith papers cite this work. Polarity classification is still indexing.
9
Pith papers citing it
years
2026 9verdicts
UNVERDICTED 9representative citing papers
GLAM learns a shared latent action space grounded in consistent future observation prediction across heterogeneous data sources to train improved behavioral cloning policies for robot manipulation tasks.
RECENT decouples skill semantics from embodiment-specific bindings via code refactoring to let small language models achieve skill grounding performance matching large language model baselines.
PHASOR factorizes motion into an FFT-based phase manifold and pose branch with semantic distillation to produce a cross-embodiment, human-anchored action embedding space for humanoid robots.
HARP aligns human-robot visual and latent action representations via paired bridges and unpaired dynamics supervision to boost VLA policy performance on manipulation tasks.
LACE aligns human-robot visual features via semantic distribution matching on corresponding body parts plus Gram loss, yielding 65% better zero-shot policy transfer than baseline DINO.
DiLA uses content-structure disentanglement driven by predictive bottlenecks to create semantically structured latent actions for high-fidelity video world models.
A dual-contrastive disentanglement method factorizes videos into independent task and embodiment latents, then uses a parameter-efficient adapter on a frozen video diffusion model to synthesize robot executions from single human demonstrations without paired data.
CORE extracts visual goal prototypes from terminal embeddings in action-free demonstrations to condition robot policies, reporting success rate gains of up to 17 percentage points on manipulation benchmarks.
citing papers explorer
-
UniLACT: Depth-Aware RGB Latent Action Learning for Vision-Language-Action Models
UniLACT improves VLA models by adding depth-aware unified latent action pretraining that outperforms RGB-only baselines on seen and unseen manipulation tasks.
-
Imitation from Heterogeneous Demonstrations using Grounded Latent-Action World Models
GLAM learns a shared latent action space grounded in consistent future observation prediction across heterogeneous data sources to train improved behavioral cloning policies for robot manipulation tasks.
-
Efficient Skill Grounding via Code Refactoring with Small Language Models
RECENT decouples skill semantics from embodiment-specific bindings via code refactoring to let small language models achieve skill grounding performance matching large language model baselines.
-
PHASOR: Phase-Anchored Universal Action Representations for Humanoid Embodiments
PHASOR factorizes motion into an FFT-based phase manifold and pose branch with semantic distillation to produce a cross-embodiment, human-anchored action embedding space for humanoid robots.
-
HARP-VLA: Human-Robot Aligned Representation Learning for Vision-Language-Action Model
HARP aligns human-robot visual and latent action representations via paired bridges and unpaired dynamics supervision to boost VLA policy performance on manipulation tasks.
-
LACE: Latent Visual Representation for Cross-Embodiment Learning
LACE aligns human-robot visual features via semantic distribution matching on corresponding body parts plus Gram loss, yielding 65% better zero-shot policy transfer than baseline DINO.
-
DiLA: Disentangled Latent Action World Models
DiLA uses content-structure disentanglement driven by predictive bottlenecks to create semantically structured latent actions for high-fidelity video world models.
-
Bridging the Embodiment Gap: Disentangled Cross-Embodiment Video Editing
A dual-contrastive disentanglement method factorizes videos into independent task and embodiment latents, then uses a parameter-efficient adapter on a frozen video diffusion model to synthesize robot executions from single human demonstrations without paired data.
-
CORE: Common Outcome Regularities from Action-Free Visual Demonstrations for Robot Manipulation
CORE extracts visual goal prototypes from terminal embeddings in action-free demonstrations to condition robot policies, reporting success rate gains of up to 17 percentage points on manipulation benchmarks.