SPECTRA-VLA: Base-Fine Hyperspectral Token Transfer at Frozen Vision-Language-Action Interfaces

cs.CV

Submitted: 2026-10-14

Updated: 2026-10-14

Comments: 34 pages: 11-page main paper and 23-page supplementary material

Code: https://github.com/jesse1029/SPECTRA-VLA

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Sources

• PointACT: Vision-Language-Action Models with Multi-Scale Point-Action Interaction
• OmniVLA: Physically-Grounded Multimodal VLA with Unified Multi-Sensor Perception for Robotic Manipulation
• MuseVLA: An Adaptive Multimodal Sensing Vision-Language-Action Model for Robotic Manipulation
• Tactile Modality Fusion for Vision-Language-Action Models
• TacVLA: Contact-Aware Tactile Fusion for Robust Vision-Language-Action Manipulation
• TAP-VLA: Tactile Annotation Prompting for Vision Language Action Models
• Neural Plasticity-Inspired Multimodal Foundation Model for Earth Observation
• HyperVision: A Channel-Adaptive Ground-Based Hyperspectral Vision Pre-trained Backbone
• Bidirectional Cross-Attention Fusion of High-Resolution RGB and Low-Resolution Hyperspectral Inputs for Multimodal Semantic Segmentation
• BridgeData V2: A Dataset for Robot Learning at Scale
• DROID: A Large-Scale In-The-Wild Robot Manipulation Dataset

Related papers

• Loss Knows Best: Detecting Annotation Errors in Videos via Loss Trajectories
• AnchorWeave: World-Consistent Video Generation with Retrieved Local Spatial Memories
• Benchmarking the Robustness of Foundation Models for Mammography under Domain Shift
• MambaX-Net: Dual-Input Mamba-Enhanced Cross-Attention Network for Longitudinal MRI Segmentation
• TeleOCR: Navigating Document Parsing Across Digital and Camera-Captured Documents
• A Survey on Efficient Vision-Language-Action Models