Research map / Vision-language and multimodal / Robot action and navigation
PRET: Planning with Directed Fidelity Trajectory for Vision and Language Navigation
Renjie Lu, Jingke Meng, WEI-SHI ZHENG
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 |
|---|---|---|---|
| represent paths with directed edge features instead of undirected node features (ablation) | PRET · R2R val unseen | SPL: 56.77 -> 65.16 | Improved |
| represent paths with directed edge features instead of undirected node features (ablation) | PRET · R2R val unseen | SR: 68.28 -> 73.78 | Improved |
| add Orientation-aware Panorama Encoder (OPE) (ablation) | PRET · R2R val unseen | SPL: 57.32 -> 65.16 | Improved |
| add Candidate Comparison Module (CCM) (ablation) | PRET · R2R val unseen | SPL: 63.07 -> 65.16 | Improved |
| add OPE and CCM on top of MAM only (ablation) | PRET · R2R val unseen | SPL: 54.48 -> 65.16 | Improved |
| use PRET planning instead of BEVBert for compute | PRET · R2R val unseen | GFLOPs at 10 steps: 35.5 -> 0.9 | Improved |
| use PRET versus BEVBert on RxR | PRET · RxR val unseen | sDTW: 58.6 -> 62.4 | Improved |
| use PRET versus BEVBert on R2R test unseen SPL | PRET · R2R test unseen | SPL: 62 -> 64 | Improved |
| use PRET versus Meta-Explore on R2R val seen | PRET · R2R val seen | SPL: 75 -> 72 | Hurt |
| use DINOv2 features instead of CLIP features (ablation) | PRET · R2R test unseen | SPL: 62 -> 64 | Improved |
Later papers that compared with this method
1 later accepted paper reports PRET as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- Bootstrapping Language-Guided Navigation Learning with Self-Refining Data Flywheel (ICLR 2025 · better on 1 of 1 shared result)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 3 on most shared results.
- DUET — DUET: Cross-Modal Semantic Grounding for Contrastive Zero-Shot Learning (AAAI 2023) · better on 2 of 2 shared results
- Meta-Explore — Meta-Explore: Exploratory Hierarchical Vision-and-Language Navigation Using Scene Object Spectrum Grounding (CVPR 2023) · better on 0 of 1 shared result
- GridMM — GridMM: Grid Memory Map for Vision-and-Language Navigation (ICCV 2023) · better on 1 of 1 shared result
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
- NavGPT: Explicit Reasoning in Vision-and-Language Navigation with Large Language Models (AAAI 2024 · similarity 0.90)
- Scaling Data Generation in Vision-and-Language Navigation (ICCV 2023 · similarity 0.89)
- Improving Vision-and-Language Navigation by Generating Future-View Image Semantics (CVPR 2023 · similarity 0.89)
- STRIDER: Navigation via Instruction-Aligned Structural Decision Space Optimization (NeurIPS 2025 · similarity 0.88)
- Affordances-Oriented Planning Using Foundation Models for Continuous Vision-Language Navigation (AAAI 2025 · similarity 0.88)
- Scene Map-based Prompt Tuning for Navigation Instruction Generation (CVPR 2025 · 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.
