Research map / Vision-language and multimodal / Robot action and navigation
SOAT: A Scene- and Object-Aware Transformer for Vision-and-Language Navigation
Read the paperCheck 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 |
|---|---|---|---|
| add object features as extra input with no architectural change (ablation) | RecBERT · R2R val-unseen | SPL: 51.43 -> 50.96 | Hurt |
| add object features with view aggregation (ablation) | RecBERT · R2R val-unseen | SPL: 50.96 -> 51.73 | Helped |
| add object features with selective object attention (ablation) | RecBERT · R2R val-unseen | SPL: 50.96 -> 52.46 | Helped |
| combine object features, view aggregation and selective attention (ablation) | RecBERT · R2R val-unseen | SPL: 51.43 -> 53.24 | Helped |
| selective attention without object features (ablation) | RecBERT · R2R val-unseen | SPL: 51.43 -> 52.06 | Helped |
| use object features with SOAT on RxR instead of scene features only | RecBERT · RxR val-unseen | SR: 40.5 -> 44.2 | Helped |
| use object features with SOAT on R2R instead of scene features only | RecBERT · R2R val-unseen | SPL: 51 -> 53 | Helped |
| remove vision-and-language pretraining (ablation) | SOAT · R2R val-unseen | SPL: 45.53 -> 44.69 | No clear effect |
| evaluate on object-heavy RxR instructions | SOAT · RxR val-unseen | SR: 33.89 -> 41.80 | Helped |
| evaluate on not-object-heavy RxR instructions | SOAT · RxR val-unseen | SR: 58.21 -> 61.64 | Helped |
| selective scene attention instead of selective object attention (ablation) | SOAT · R2R val-unseen | SPL: 53.24 -> 50.99 | Hurt |
| all attention instead of selective object attention (ablation) | SOAT · R2R val-unseen | SPL: 53.24 -> 51.73 | Hurt |
Closest papers by meaning
- Towards Learning a Generic Agent for Vision-and-Language Navigation via Pre-Training (CVPR 2020 · similarity 0.90)
- Meta-Explore: Exploratory Hierarchical Vision-and-Language Navigation Using Scene Object Spectrum Grounding (CVPR 2023 · similarity 0.90)
- General Scene Adaptation for Vision-and-Language Navigation (ICLR 2025 · similarity 0.90)
- LangMap: A Human-Verified Benchmark for Hierarchical Open-Vocabulary Goal Navigation (NeurIPS 2026 · similarity 0.89)
- NavGPT: Explicit Reasoning in Vision-and-Language Navigation with Large Language Models (AAAI 2024 · similarity 0.88)
- Hierarchical Semantic-Augmented Navigation: Optimal Transport and Graph-Driven Reasoning for Vision-Language Navigation (NeurIPS 2025 · similarity 0.87)
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.