Research map / Vision-language and multimodal / Robot action and navigation
Think Global, Act Local: Dual-Scale Graph Transformer for Vision-and-Language Navigation
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 dual-scale map and local visual representations with dynamic fusion rather than either scale alone (ablation) | DUET · REVERIE val unseen | SPL: 25.98 -> 33.73 | Helped |
| Replace average fusion with dynamic fusion (ablation) | DUET · REVERIE val unseen | SPL: 31.94 -> 33.73 | Helped |
| Add graph-aware self-attention to dynamic fusion (ablation) | DUET · REVERIE val unseen | SPL: 32.88 -> 33.73 | Helped |
| Add object grounding supervision to SAP training (ablation) | DUET · REVERIE val unseen | SR: 35.30 -> 37.80 | Helped |
| Add MLM and MRC proxy tasks to SAP + OG training (ablation) | DUET · REVERIE val unseen | RGS: 23.89 -> 25.05 | Helped |
| Fine-tune with PID rather than reinforcement learning (ablation) | DUET · REVERIE val unseen | SR: 42.35 -> 46.98 | Helped |
| Add synthetic instructions during pretraining without PID (ablation) | DUET · REVERIE val unseen | SPL: 25.56 -> 27.19 | Helped |
| Add synthetic instruction augmentation when PID is used (ablation) | DUET · REVERIE val unseen | SPL: 33.73 -> 32.72 | Hurt |
| Increase behavior-cloning balance factor from 0.2 to 1 (ablation) | DUET · REVERIE val unseen | SR: 46.98 -> 45.64 | Hurt |
| Remove behavior cloning from fine-tuning (λ = 0) (ablation) | DUET · REVERIE val unseen | RGSPL: 23.03 -> 22.04 | Hurt |
| Use DUET instead of HAMT | REVERIE val unseen · REVERIE val unseen | SR: 32.95 -> 46.98 | Helped |
| Use DUET instead of GBE | SOON test unseen · SOON test unseen | SR: 12.90 -> 33.44 | Helped |
Closest papers by meaning
- NavGPT: Explicit Reasoning in Vision-and-Language Navigation with Large Language Models (AAAI 2024 · similarity 0.89)
- Hierarchical Semantic-Augmented Navigation: Optimal Transport and Graph-Driven Reasoning for Vision-Language Navigation (NeurIPS 2025 · similarity 0.89)
- General Scene Adaptation for Vision-and-Language Navigation (ICLR 2025 · similarity 0.88)
- NavBench: Probing Multimodal Large Language Models for Embodied Navigation (NeurIPS 2025 · similarity 0.88)
- Human-Aware Vision-and-Language Navigation: Bridging Simulation to Reality with Dynamic Human Interactions (NeurIPS 2024 · similarity 0.88)
- D3D-VLP: Dynamic 3D Vision-Language-Planning Model for Embodied Grounding and Navigation (CVPR 2026 · similarity 0.88)
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.