Research map / Vision-language and multimodal / Robot action and navigation
LatentPilot: Scene-Aware Vision-and-Language Navigation by Dreaming Ahead with Latent Visual Reasoning
HAIHONG HAO, Lei Chen, Mingfei Han, Changlin Li, Dong An, Yuqiang Yang, Zhihui Li, Xiaojun Chang
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 |
|---|---|---|---|
| supervise Pilot Token with future visual latents instead of no extra supervision (ablation) | LatentPilot · R2R Val-Unseen | SR: 51.7 -> 62.0 | Improved |
| supervise Pilot Token with future visual latents instead of 3D features (ablation) | LatentPilot · R2R Val-Unseen | SR: 50.3 -> 62.0 | Improved |
| supervise Pilot Token with future visual latents instead of text (ablation) | LatentPilot · R2R Val-Unseen | SR: 53.2 -> 62.0 | Improved |
| add 3D (VGGT) supervision to Pilot Token (ablation) | LatentPilot · R2R Val-Unseen | SR: 51.7 -> 50.3 | Hurt |
| add text supervision to Pilot Token (ablation) | LatentPilot · R2R Val-Unseen | SR: 51.7 -> 53.2 | No clear effect |
| add future vision supervision to Pilot Token (ablation) | LatentPilot · R2R Val-Seen | SR: 53.1 -> 65.1 | Improved |
| internalize future supervision instead of plug-in Wan2.1 world model (ablation) | LatentPilot · R2R Val-Seen | SR: 56.6 -> 65.1 | Improved |
| internalize future supervision instead of plug-in CogVideoX1.5 world model (ablation) | LatentPilot · R2R Val-Seen | SR: 59.3 -> 65.1 | Improved |
| internalize future supervision instead of plug-in Wan2.1 world model (ablation) | LatentPilot · R2R Val-Seen | time per action (ms): 2040 -> 130 | Improved |
| add more PilotLoop flywheel rounds (ablation) | LatentPilot · R2R Val-Unseen | SR and SPL | Improved |
| LatentPilot vs JanusVLN | LatentPilot · R2R-CE Val-Unseen | SR: 60.5 -> 62.0 | Improved |
| LatentPilot vs JanusVLN | LatentPilot · RxR-CE Val-Unseen | SR: 56.2 -> 58.2 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- JanusVLN — JanusVLN: Decoupling Semantics and Spatiality with Dual Implicit Memory for Vision-Language Navigation (ICLR 2026) · better on 4 of 4 shared results
- ETPNav — ETPNav: Evolving Topological Planning for Vision-Language Navigation in Continuous Environments (TPAMI 2024) · better on 2 of 2 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Toward a Causally-Inspired Evolving World Model for Vision-and-Language Navigation in Continuous Environments (TPAMI 2026 · similarity 0.92)
- A Latent World-Action Model with Jointly Aligned Reasoning (NeurIPS 2026 · similarity 0.91)
- Reasoning in Visual Navigation of End-to-end Trained Agents: A Dynamical Systems Approach (CVPR 2025 · similarity 0.91)
- FantasyVLN: Unified Multimodal Chain-of-Thought Reasoning for Vision-and-Language Navigation (CVPR 2026 · similarity 0.90)
- NavGPT: Explicit Reasoning in Vision-and-Language Navigation with Large Language Models (AAAI 2024 · similarity 0.90)
- OneVL: One-Step Latent Reasoning and Planning with Vision-Language Explanation (NeurIPS 2026 · 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.
