Research map / Multi-agent systems and games / Human-AI and goal-conditioned
Improving Vision-and-Language Navigation by Generating Future-View Image Semantics
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 |
|---|---|---|---|
| Combine MTM, MPM, and APIG pre-training (ablation) | HAMT · R2R Val Unseen | SR: 64.4 -> 68.1 | Helped |
| Add MTM pre-training (ablation) | HAMT · R2R Val Unseen | SR: 64.4 -> 66.7 | Helped |
| Add MPM pre-training (ablation) | HAMT · R2R Val Unseen | SR: 64.4 -> 65.5 | Helped |
| Add APIG pre-training (ablation) | HAMT · R2R Val Unseen | SR: 64.4 -> 66.3 | Helped |
| Use dynamic codebook selection instead of no codebook selection for MTM (ablation) | MTM · R2R Val Unseen | SR: 65.1 -> 66.7 | Helped |
| Use dynamic codebook selection instead of no codebook selection for APIG (ablation) | APIG · R2R Val Unseen | SR: 64.4 -> 66.3 | Helped |
| Use static codebook selection for MPM instead of no selection (ablation) | MPM · R2R Val Unseen | SR: 64.8 -> 65.5 | Helped |
| Replace mean patch probability with sampled patch probability (ablation) | VLN-SIG · R2R Val Unseen | SR: 65.5 -> 65.9 | Helped |
| Replace mean patch probability with weighted patch probability (ablation) | VLN-SIG · R2R Val Unseen | SR: 65.5 -> 66.9 | Helped |
| Use block-wise sampling for weighted patch probability (ablation) | VLN-SIG · R2R Val Unseen | SR: 66.9 -> 68.1 | Helped |
| Add APIG as a fine-tuning auxiliary task (ablation) | VLN-SIG with static codebook · R2R Val Unseen | SR: 64.9 -> 65.9 | Helped |
| Add APIG as a fine-tuning auxiliary task (ablation) | VLN-SIG with dynamic codebook · R2R Val Unseen | SR: 66.1 -> 66.3 | Helped |
Closest papers by meaning
- Towards Learning a Generic Agent for Vision-and-Language Navigation via Pre-Training (CVPR 2020 · similarity 0.93)
- Generative Language-Grounded Policy in Vision-and-Language Navigation with Bayes' Rule (ICLR 2021 · similarity 0.91)
- History to Future: Evolving Agent with Experience and Thought for Zero-shot Vision-and-Language Navigation (CVPR 2026 · similarity 0.91)
- NavGPT: Explicit Reasoning in Vision-and-Language Navigation with Large Language Models (AAAI 2024 · similarity 0.90)
- Lookahead Exploration with Neural Radiance Representation for Continuous Vision-Language Navigation (CVPR 2024 · similarity 0.90)
- Do Visual Imaginations Improve Vision-and-Language Navigation Agents? (CVPR 2025 · similarity 0.89)
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.