VLA models from VLM adaptation can be pruned 12-30% via multi-module joint scheme based on divergence signals while keeping ~90% performance on LIBERO without post-pruning recovery, unlike standard criteria that collapse.
arXiv preprint arXiv:2505.21200 (2025)
13 Pith papers cite this work. Polarity classification is still indexing.
13
Pith papers citing it
citation-role summary
background 3
citation-polarity summary
roles
background 3polarities
background 3representative citing papers
VLA language backbones show high redundancy on manipulation benchmarks, with half the LLM blocks removable and even two blocks sufficient to recover baseline performance after fine-tuning, unlike vision and action pathways.
Pace-and-Path Correction decomposes a quadratic cost minimization into orthogonal pace and path channels to correct chunked actions in VLA models, raising success rates by up to 28.8% in dynamic settings.
KERV integrates kinematic Kalman Filter predictions with speculative decoding in VLA models to achieve 27-37% faster inference while maintaining nearly the same task success rates.
ISR resamples trajectories to equal information spacing via velocity/acceleration norms on a Riemannian manifold, raising imitation learning success rates by ~25% over time-uniform downsampling on three real-world manipulation tasks.
UniFS achieves 98.3% success on LIBERO with 2.1x lower latency than prior fast-slow VLA models by stratifying VLM layer update frequencies, inverting latent interactions, and applying multi-level supervision.
VLA-InfoEntropy accelerates Vision-Language-Action model inference by using visual entropy, attention entropy, and timestep cues to prune redundant tokens while preserving task-critical content.
OxyGen unifies KV cache management in MoT VLAs to enable cross-task KV sharing and cross-frame continuous batching, delivering up to 3.7x speedup with 200+ tokens/s language and 70 Hz action on on-device platforms.
ActDistill transfers action knowledge from heavy VLA teacher models to lightweight students via graph-encapsulated hierarchies and action-guided dynamic routing, delivering over 50% computation reduction and 1.67x speedup with comparable or better performance on embodied tasks.
GeoSem-WAM adds geometric and semantic auxiliary prediction tasks to World Action Models during training to improve latent representations and action prediction accuracy while keeping inference efficient by avoiding explicit future rollouts.
ElegantVLA accelerates VLA models up to 3.77x by dynamically scheduling compute across vision, language, and action components without retraining the base model.
This survey organizes large VLM-based VLA models for robotic manipulation into monolithic and hierarchical paradigms, reviews their integrations and datasets, and outlines future directions.
AttenA+ reweights action training objectives in VLA and WAM models via inverse velocity attention to prioritize kinematically critical segments, yielding small benchmark gains.
citing papers explorer
No citing papers match the current filters.