Research map / Vision-language and multimodal / Robot action and navigation
Sim-2-Sim Transfer for Vision-and-Language Navigation in Continuous Environments
Jacob Krantz, Stefan Lee
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 |
|---|---|---|---|
| train VLN-BERT on VLN-CE reconstructed renders instead of MP3D panoramas (ablation) | VLN-BERT · VLN Val-Unseen evaluated on Recon. | SR: 58 -> 60 | Improved |
| train on reconstructed renders instead of MP3D panoramas (ablation) | VLN-BERT · VLN Val-Seen evaluated on Recon. | SR: 62 -> 72 | Improved |
| evaluate on reconstructed renders instead of MP3D panoramas without retraining | VLN-BERT · VLN Train* split | SR: 93 -> 80 | Hurt |
| use VLN-CE subset instead of full R2R | VLN-BERT · VLN Val-Seen | SR: 72 -> 69 | Hurt |
| use Oracle Policy instead of teleportation (ablation) | VLN-BERT · VLN-CE Val-Unseen | SR: 60 -> 60 | No clear effect |
| use realistic Local Policy instead of Oracle Policy (ablation) | VLN-BERT · VLN-CE Val-Unseen | SR: 60 -> 55 | Hurt |
| use optimal SGM output space instead of nav-graph candidates (ablation) | VLN-BERT · VLN-CE Val-Unseen | SR: 55 -> 49 | Hurt |
| use predicted SGM candidates instead of optimal SGM candidates (ablation) | VLN-BERT · VLN-CE Val-Unseen | SR: 49 -> 41 | Hurt |
| fine-tune VLN-BERT on SGM candidates (ablation) | VLN-BERT · VLN-CE Val-Unseen | SR: 41 -> 43 | Improved |
| train SGM with Habitat-rendered vision instead of MP3D panoramas (ablation) | SGM · VLN-CE Val-Unseen | SR: 38 -> 41 | Improved |
| use 360 degree laser scan instead of 270 degree (ablation) | SGM · VLN-CE Val-Unseen | SR: 33 -> 38 | Improved |
| transfer VLN-BERT to VLN-CE with the harness instead of training end-to-end in VLN-CE | VLN-CE-BERT · VLN-CE Test | SR: 32 -> 44 | Improved |
Closest papers by meaning
- Iterative Vision-and-Language Navigation (CVPR 2023 · similarity 0.90)
- Bridging the Gap Between Learning in Discrete and Continuous Environments for Vision-and-Language Navigation (CVPR 2022 · similarity 0.90)
- Beyond the Nav-Graph: Vision-and-Language Navigation in Continuous Environments (ECCV 2020 · similarity 0.89)
- STRIDER: Navigation via Instruction-Aligned Structural Decision Space Optimization (NeurIPS 2025 · similarity 0.88)
- General Scene Adaptation for Vision-and-Language Navigation (ICLR 2025 · similarity 0.88)
- Human-Aware Vision-and-Language Navigation: Bridging Simulation to Reality with Dynamic Human Interactions (NeurIPS 2024 · 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.
