Introduces LIBERO-Occ benchmark showing VLA performance drop under occlusion and Viewpoint Imagination method that generates complementary views to improve robustness without extra hardware.
Stereovla: Enhancing vision- language-action models with stereo vision
8 Pith papers cite this work. Polarity classification is still indexing.
8
Pith papers citing it
abstract
While Vision-Language-Action (VLA) models excel in generalist manipulation, they often lack fine-grained spatial awareness and show limited viewpoint robustness. This limitation largely stems from the reliance on pretrained RGB encoders, which lack explicit geometric cues and prioritize semantic alignment over geometric representation. We argue that effective visual representations for VLA models must jointly encode both semantic and geometric information. In this paper, we introduce StereoVLA, the first VLA model to incorporate rich geometric cues from large-scale synthetic stereo data. StereoVLA employs a Geometric-and-Semantic (GeoSem) vision encoder that extracts geometric cues from subtle stereo-view disparities for precise spatial perception, while simultaneously capturing semantic features from pixel observations to support language-conditioned manipulation. Additionally, we introduce two synergistic co-training objectives: Interaction-Region Depth Estimation for precise spatial reasoning, and Camera Parameter Estimation to implicitly align perception and action coordinate systems. Compared with baselines that employ various input modalities, StereoVLA achieves a 33.4% absolute gain in success rate in real-world experiments and demonstrates robustness to near-hemispheric camera perspectives. Project page: https://shengliangd.github.io/StereoVLA-Webpage.
citation-role summary
background 2
method 1
citation-polarity summary
years
2026 8representative citing papers
CamVLA decouples robot actions into camera-frame movements and a learned hand-eye pose, enabling calibration-free manipulation under unseen camera viewpoints from a single RGB image.
Dexterity-BEV creates 3D vertex-based inputs and BEV-aligned outputs to reduce spatial-temporal misalignments in end-to-end robot policies trained on diverse datasets and embodiments.
GuidedVLA improves VLA generalization by supervising individual attention heads with manually defined auxiliary signals for three task-relevant factors.
MolmoAct2 is an open VLA model that outperforms baselines like Pi-05 on 7 benchmarks and whose backbone surpasses GPT-5 on 13 embodied-reasoning tasks through new datasets, specialized training, and architecture changes for lower latency.
Event-VLA integrates event streams into VLA models through action-conditioned gated cross-attention to maintain performance in normal light while improving success rates under low-light and near-dark conditions.
Degradation-Driven Prompting improves VLM accuracy on deceptive visual tasks by downsampling inputs and routing them through classification, external visual tools, and a critic.
Evo-Depth is a compact VLA model using a lightweight implicit depth encoder from RGB views plus progressive alignment to boost manipulation performance without added hardware.
citing papers explorer
-
LIBERO-Occ: Evaluating and Improving Vision-Language-Action Models under Scene-Induced Occlusion via Viewpoint Imagination
Introduces LIBERO-Occ benchmark showing VLA performance drop under occlusion and Viewpoint Imagination method that generates complementary views to improve robustness without extra hardware.
-
From Fixed to Free Cameras: Calibration-Free View-Robust Vision-Language-Action Model
CamVLA decouples robot actions into camera-frame movements and a learned hand-eye pose, enabling calibration-free manipulation under unseen camera viewpoints from a single RGB image.
-
Dexterity-BEV: Aligning 3D World and Actions for Generalizable Robot Policies Learning
Dexterity-BEV creates 3D vertex-based inputs and BEV-aligned outputs to reduce spatial-temporal misalignments in end-to-end robot policies trained on diverse datasets and embodiments.
-
GuidedVLA: Specifying Task-Relevant Factors via Plug-and-Play Action Attention Specialization
GuidedVLA improves VLA generalization by supervising individual attention heads with manually defined auxiliary signals for three task-relevant factors.
-
MolmoAct2: Action Reasoning Models for Real-world Deployment
MolmoAct2 is an open VLA model that outperforms baselines like Pi-05 on 7 benchmarks and whose backbone surpasses GPT-5 on 13 embodied-reasoning tasks through new datasets, specialized training, and architecture changes for lower latency.
-
Event-VLA: Action-Conditioned Event Fusion for Robust Vision-Language-Action Model
Event-VLA integrates event streams into VLA models through action-conditioned gated cross-attention to maintain performance in normal light while improving success rates under low-light and near-dark conditions.
-
E-VLA: Event-Augmented Vision-Language-Action Model for Dark and Blurred Scenes
Degradation-Driven Prompting improves VLM accuracy on deceptive visual tasks by downsampling inputs and routing them through classification, external visual tools, and a critic.
-
Evo-Depth: A Lightweight Depth-Enhanced Vision-Language-Action Model
Evo-Depth is a compact VLA model using a lightweight implicit depth encoder from RGB views plus progressive alignment to boost manipulation performance without added hardware.