Research map / Vision-language and multimodal / Robot action and navigation
Zero-shot Active Mapping via Fused 360-BEV Representations and Vision–Language Models
Yuanze Wang, Dianxi Shi, Yuetian Wang, Shiming Song, Haikuo Peng, Chunping Qiu, Mengzhu Wang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use fused 360-BEV representation instead of pinhole-BEV (ablation) | VLM active mapping agent · GLEAM-Bench overall | Coverage: 38.77% -> 75.31% | Improved |
| have VLM select numbered image-space candidate waypoints instead of predicting BEV coordinates (ablation) | VLM active mapping agent · GLEAM-Bench overall | Coverage: 45.10% -> 75.31% | Improved |
| use a larger VLM (Qwen3-VL-235B-A22B vs 8B) (ablation) | VLM active mapping agent · GLEAM-Bench overall | Coverage: 50.35% -> 75.31% | Improved |
| use VLM waypoint selection instead of Nearest-to-Centroid heuristic (ablation) | VLM active mapping agent · GLEAM-Bench overall | Coverage: 47.96% -> 75.31% | Improved |
| use VLM waypoint selection instead of Max-Frontier-Distance heuristic (ablation) | VLM active mapping agent · GLEAM-Bench overall | Coverage: 52.85% -> 75.31% | Improved |
| use depth-first rather than breadth-first exploration (ablation) | VLM active mapping agent · GLEAM-Bench overall | Trajectory length: 82.77m -> 52.03m | Improved |
| use depth-first rather than breadth-first exploration (ablation) | VLM active mapping agent · GLEAM-Bench overall | Coverage: 73.82% -> 75.31% | No clear effect |
| zero-shot VLM agent instead of trained RL method GLEAM | active mapping · GLEAM-Bench overall | Coverage: 66.50% -> 75.31% | Improved |
Closest papers by meaning
- Agent3D-Zero: An Agent for Zero-shot 3D Understanding (ECCV 2024 · similarity 0.89)
- IR3D-Bench: Evaluating Vision-Language Model Scene Understanding as Agentic Inverse Rendering (NeurIPS 2025 · similarity 0.88)
- D3D-VLP: Dynamic 3D Vision-Language-Planning Model for Embodied Grounding and Navigation (CVPR 2026 · similarity 0.87)
- BeliefMapNav: 3D Voxel-Based Belief Map for Zero-Shot Object Navigation (NeurIPS 2025 · similarity 0.87)
- Affordances-Oriented Planning Using Foundation Models for Continuous Vision-Language Navigation (AAAI 2025 · similarity 0.87)
- Human-Aware Vision-and-Language Navigation: Bridging Simulation to Reality with Dynamic Human Interactions (NeurIPS 2024 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
