Research map / Vision-language and multimodal / Robot action and navigation
NavQ: Learning a Q-Model for Foresighted Vision-and-Language Navigation
Peiran Xu, Xicheng Gong, Yadong Mu
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 |
|---|---|---|---|
| add Q-model and future encoder (ablation) | DUET baseline · REVERIE val unseen | SR: 48.14 -> 53.22 | Improved |
| add future encoder only (ablation) | DUET baseline · REVERIE val unseen | SR: 48.14 -> 48.20 | No clear effect |
| add Q-model only (concatenated to GE) (ablation) | DUET baseline · REVERIE val unseen | SR: 48.14 -> 48.48 | No clear effect |
| add progress-related subtask losses (ablation) | NavQ · REVERIE val unseen | SR: 51.55 -> 53.22 | Improved |
| add progress-related subtask losses (ablation) | NavQ · REVERIE val unseen | RGSPL: 23.81 -> 27.12 | Improved |
| replace predicted Q-features with ground truth (ablation) | NavQ · REVERIE val unseen | SR: 53.22 -> 59.27 | Improved |
| set decay ratio 0.5 instead of 0 (ablation) | NavQ · REVERIE val unseen | SR: 51.12 -> 53.22 | Improved |
| set decay ratio 0.5 instead of 0.7 (ablation) | NavQ · REVERIE val unseen | SR: 50.89 -> 53.22 | Improved |
| set decay ratio 0.5 instead of 0.3 (ablation) | NavQ · REVERIE val unseen | SPL: 38.90 -> 38.89 | No clear effect |
| add unannotated HM3D/Gibson scenes for Q-model training | NavQ · REVERIE val unseen | SR: 53.22 -> 54.10 | Improved |
| add NavQ to DUET | DUET · REVERIE test unseen | SPL: 36.06 -> 39.50 | Improved |
| add NavQ to DUET | DUET · SOON val unseen | SR: 36.28 -> 39.09 | Improved |
Closest papers by meaning
- NavCoT: Boosting LLM-Based Vision-and-Language Navigation via Learning Disentangled Reasoning (TPAMI 2025 · similarity 0.89)
- Improving Vision-and-Language Navigation by Generating Future-View Image Semantics (CVPR 2023 · similarity 0.89)
- NavGPT: Explicit Reasoning in Vision-and-Language Navigation with Large Language Models (AAAI 2024 · similarity 0.89)
- NavBench: Probing Multimodal Large Language Models for Embodied Navigation (NeurIPS 2025 · similarity 0.87)
- Towards Long-Horizon Vision-Language Navigation: Platform, Benchmark and Method (CVPR 2025 · similarity 0.87)
- Towards Realistic UAV Vision-Language Navigation: Platform, Benchmark, and Methodology (ICLR 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.
