Spatial-Semantic Uncertainty in VLM-Based Target Search: Balancing Exploration and Identification

arXiv:2609.20443 · cs.RO · Submitted 2026-09-17 · Read on arXiv

cs.RO

Submitted: 2026-09-17

Updated: 2026-09-17

Comments: Accepted for presentation at IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2026) WorkshopRethinking Uncertainty for Modern Robotics Paradigms

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

Terminology

• Accuracy — 12× in this paper · explained in 1 episode
• Calibration — 7× in this paper · explained in 7 episodes
• Information Gain — 3× in this paper · explained in 2 episodes
• Consistency — 2× in this paper · explained in 3 episodes
• Additive Decomposition — 1× in this paper · explained in 1 episode
• Benchmark — 1× in this paper · explained in 2 episodes
• Epistemic Uncertainty — 1× in this paper · explained in 9 episodes
• Motion Primitives — 1× in this paper · explained in 2 episodes
• Vision-Language Models (VLMs) — 1× in this paper · explained in 6 episodes

Sources

• Explore until Confident: Efficient Exploration for Embodied Question Answering
• VLAConf: Calibrated Task-Success Confidence for Vision-Language-Action Models
• IGV-RRT: Prior-Real-Time Observation Fusion for Active Object Search in Changing Environments

Related papers

• FMT x: An Efficient and Asymptotically Optimal Extension of the Fast Marching Tree for Dynamic Replanning
• MPCFormer: A physics-informed data-driven approach for explainable socially-aware autonomous driving
• RoboLab: A High-Fidelity Simulation Benchmark for Analysis of Task Generalist Policies
• ARC: A Reasoning Recipe for Robot Foundation Models
• HRDexDB: A 4D Dexterous Grasping Dataset Across Human and Multiple Robot Embodiments
• APT: Action Expert Pretraining Improves Instruction Generalization of Vision-Language-Action Policies