Research map / Vision-language and multimodal / Robot action and navigation
Counterfactual Vision-and-Language Navigation via Adversarial Path Sampler
Tsu-Jui Fu, Xin Eric Wang, Matthew F. Peterson, Scott T. Grafton, Miguel P. Eckstein, William Yang Wang
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 APS-sampled instead of randomly-sampled augmented paths | Seq2Seq · R2R val-seen | Success Rate (SR): 43.7 -> 48.2 | Improved |
| add APS-sampled instead of randomly-sampled augmented paths | Speaker-Follower · R2R val-seen | Success Rate (SR): 66.4 -> 68.2 | Improved |
| add APS-sampled instead of randomly-sampled augmented paths | RCM · R2R val-seen | Success Rate (SR): 61.9 -> 63.2 | Improved |
| add APS-sampled instead of randomly-sampled augmented paths | Seq2Seq · R2R val-unseen | Success Rate (SR): 22.6 -> 24.2 | Improved |
| add APS-sampled instead of randomly-sampled augmented paths | Speaker-Follower · R2R val-unseen | Success Rate (SR): 36.1 -> 38.8 | Improved |
| add APS-sampled instead of randomly-sampled augmented paths | RCM · R2R val-unseen | Success Rate (SR): 45.6 -> 47.7 | Improved |
| add APS-sampled instead of randomly-sampled augmented paths | Seq2Seq · R2R test | Success Rate (SR): 21.0 -> 22.5 | Improved |
| add APS-sampled instead of randomly-sampled augmented paths | Speaker-Follower · R2R test | Success Rate (SR): 34.8 -> 36.1 | Improved |
| add APS-sampled instead of randomly-sampled augmented paths | RCM · R2R test | Success Rate (SR): 44.5 -> 45.1 | Improved |
| add environment-based pre-exploration with APS | Seq2Seq + augaps · R2R val-unseen | Success Rate (SR): 24.2 -> 27.0 | Improved |
| add environment-based pre-exploration with APS | Speaker-Follower + augaps · R2R val-unseen | Success Rate (SR): 38.8 -> 42.0 | Improved |
| add environment-based pre-exploration with APS | RCM + augaps · R2R val-unseen | Success Rate (SR): 47.7 -> 48.0 | Improved |
Closest papers by meaning
- Counterfactual Vision-and-Language Navigation: Unravelling the Unseen (NeurIPS 2020 · similarity 0.92)
- Human-Aware Vision-and-Language Navigation: Bridging Simulation to Reality with Dynamic Human Interactions (NeurIPS 2024 · similarity 0.86)
- FantasyVLN: Unified Multimodal Chain-of-Thought Reasoning for Vision-and-Language Navigation (CVPR 2026 · similarity 0.86)
- CofactVLA: Deconfounding Vision-Language-Action Models via Counterfactual Intervention (NeurIPS 2026 · similarity 0.86)
- Run, Ruminate, and Regulate: A Dual-process Thinking System for Vision-and-Language Navigation (AAAI 2026 · similarity 0.86)
- Aux-Think: Exploring Reasoning Strategies for Data-Efficient Vision-Language Navigation (NeurIPS 2025 · similarity 0.86)
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.
