20 papers
Relational Geometry Forcing for World Action Models
Yuan Xu, Yixiang Chen, Qisen Ma +10
World Action Models (WAMs) have shown strong potential for robot manipulation by jointly learning visual dynamics and executable actions. However, their video and action objectives…
What Matters for World Modeling in Vision-Language-Action Models?
Qisen Ma, Yixiang Chen, Yuan Xu +10
Integrating world models into Vision-Language-Action models (VLAs) has emerged as a promising direction, enabling policies to anticipate future dynamics beyond reactive observation…
GuardianBench: A Same-Scene Instruction-Contrastive Benchmark for Latent Contextual Risk in Embodied AI
Zhesheng Zhang, Jiahao Lu, Wei Liu +8
In embodied AI, safety risk can be latent: a benign instruction and a safe scene become hazardous only when composed. Prior work has advanced embodied safety by varying visual cont…
XEWorld: Can Action-Conditioned World Models Generalize to Unseen Robot Embodiments?
Yixiang Chen, Jiabing Yang, Yuan Xu +10
Action-conditioned world models are promising learned simulators for robotic manipulation, yet evaluating them exclusively on training robots fails to reveal whether they capture p…
BridgeVLA++: A Data-Efficient, Generalizable, and Memory-Augmented Vision-Language-Action Framework for 3D Manipulation
Peiyan Li, Yuze Zhu, Yixiang Chen +10
Leveraging pre-trained vision-language models (VLMs) to construct vision-language-action (VLA) models has emerged as a promising paradigm for 3D robot manipulation. However, existi…
Style over Substance: A Shortcut Audit of Emotion-Description Preference Evaluation
Jiabing Yang, Yixiang Chen, Yuan Xu +6
Preference over model-generated emotion descriptions is emerging as a standard evaluation metric for multimodal emotion understanding, exemplified by the MER2026 MER-Prefer track o…
UESF-Bench: Benchmarking and Probing for Unified Embodied Seeking and Following
Kun Yu, Jianhua Yang, Yixiang Chen +7
Language-guided human following is an important capability for embodied agents, but existing benchmarks typically assume that the target person is visible at the start of an episod…
FlowWAM: Optical Flow as a Unified Action Representation for World Action Models
Yixiang Chen, Peiyan Li, Yuan Xu +13
World Action Models (WAMs) are able to leverage pretrained video generators for both world modeling and action prediction. However, directly leveraging such video generators for co…
DIM-WAM: World-Action Modeling with Diverse Historical Event Memory
Kai Wang, Zhaopeng Gu, Yixiang Chen +7
World-action models have shown promising robot-manipulation performance by jointly predicting future visual states and actions. However, existing methods mainly rely on short-term…
Improving Vision-Language-Action Model Fine-Tuning with Structured Stage and Keyframe Supervision
Yuan Xu, Yixiang Chen, Kai Wang +5
Vision-Language-Action (VLA) models have shown strong potential for generalizable robotic manipulation. During fine-tuning, however, action supervision applies equally across all t…
SKIP: Sparse Keyframe Interpolation Paradigm for Efficient Embodied World Models
Ziheng He, Yixiang Chen, Ning Yang +11
Embodied world models have emerged as a promising paradigm in robotics by predicting how robot actions affect the surrounding scene. However, the rollout inference remains computat…
SpatialVAM:Spatial-Aware Multi-View Video Diffusion as a Data-Efficient Robot Policy
Peiyan Li, Yixiang Chen, Yuan Xu +13
Robotic manipulation requires understanding both the 3D spatial structure of the environment and its temporal evolution, yet most existing policies neglect one or both aspects. The…
BridgeV2W: Bridging Video Generation Models to Embodied World Models via Embodiment Masks
Yixiang Chen, Peiyan Li, Jiabing Yang +8
Embodied world models have emerged as a promising paradigm in robotics, most of which leverage large-scale Internet videos or pretrained video generation models to enrich visual an…
UAOR: Uncertainty-aware Observation Reinjection for Vision-Language-Action Models
Jiabing Yang, Yixiang Chen, Yuan Xu +14
Vision-Language-Action (VLA) models leverage pretrained Vision-Language Models (VLMs) as backbones to map images and instructions to actions, demonstrating remarkable potential for…
VERM: Leveraging Foundation Models to Create a Virtual Eye for Efficient 3D Robotic Manipulation
Yixiang Chen, Yan Huang, Keji He +2
When performing 3D manipulation tasks, robots have to execute action planning based on perceptions from multiple fixed cameras. The multi-camera setup introduces substantial redund…
EgoDemoGen: Egocentric Demonstration Generation for Viewpoint Generalization in Robotic Manipulation
Yuan Xu, Jiabing Yang, Xiaofeng Wang +16
Imitation learning based visuomotor policies have achieved strong performance in robotic manipulation, yet they often remain sensitive to egocentric viewpoint shifts. Unlike third-…
IKOD: Mitigating Visual Attention Degradation in Large Vision-Language Models
Jiabing Yang, Chenhang Cui, Yiyang Zhou +6
Recent advancements in Large Vision-Language Models (LVLMs) have demonstrated significant progress across multiple domains. However, these models still face the inherent challenge…
LaPA: Length-Aware Prefix and Prompt Attention Augmentation for Long-Form Controllable Text Generation
Jiabing Yang, Yixiang Chen, Zichen Wen +8
Prefix-based methods have emerged as a promising paradigm for Controllable Text Generation (CTG) due to their parameter efficiency. However, while effective in short sequences, the…
EC-Flow: Enabling Versatile Robotic Manipulation from Action-Unlabeled Videos via Embodiment-Centric Flow
Yixiang Chen, Peiyan Li, Yan Huang +3
Current language-guided robotic manipulation systems often require low-level action-labeled datasets for imitation learning. While object-centric flow prediction methods mitigate t…
BridgeVLA: Input-Output Alignment for Efficient 3D Manipulation Learning with Vision-Language Models
Peiyan Li, Yixiang Chen, Hongtao Wu +6
Recently, leveraging pre-trained vision-language models (VLMs) for building vision-language-action (VLA) models has emerged as a promising approach to effective robot manipulation…