Abstract
Aerial vision-and-language navigation (VLN) agents are typically trained on detail-rich, trajectory-aligned commands, whereas users issue short, intent-driven instructions; on a frozen OpenFly navigator, this \emph{instruction gap} drops success rate (SR) from $31.03\%$ to $11.33\%$. To scale translator training, we prompt a language model with human-written style examples to convert original commands into paired, intent-centered Weak commands, which yield $15.27\%$ SR. We introduce the \textbf{Trajectory-Grounded Instruction Translator (TGIT)}, a front-end that keeps the navigator frozen and translates Weak inputs into agent-executable commands by learning from its trajectory outcomes. The resulting Weak-trained translator raises Weak-input SR to $37.93\%$ and transfers zero-shot to real human instructions ($11.33\%{\rightarrow}32.51\%$); it also improves held-out OpenFly ($4.95\%{\rightarrow}20.79\%$) and yields recovery on CityNav and AirVLN.
Keywords
Publication details
- Journal
- Not available
- Open access
- Green open access
Cite this article
APA 7
Chen, X., Liu, Z., Xu, X., Xu, J., Zhou, C., Su, Y., Zeng, R., Du, T., Yao, K., Li, C., & Ji, S. (2026). Speaking the Navigator's Language: Trajectory-Grounded Instruction Translation for Frozen Aerial VLN Agents. https://omanscience.com/en/articles/speaking-the-navigator-s-language-trajectory-grounded-instruction-translation-for-frozen-aerial-vln-agents
MLA 9
Chen, Xi, et al. "Speaking the Navigator's Language: Trajectory-Grounded Instruction Translation for Frozen Aerial VLN Agents." https://omanscience.com/en/articles/speaking-the-navigator-s-language-trajectory-grounded-instruction-translation-for-frozen-aerial-vln-agents.
Chicago (author–date)
Chen, Xi, Zhe Liu, Xiaogang Xu, Jiafei Xu, Chunyi Zhou, Yuan Su, Rui Zeng, Tianyu Du, Kelu Yao, Chao Li, and Shouling Ji. 2026. "Speaking the Navigator's Language: Trajectory-Grounded Instruction Translation for Frozen Aerial VLN Agents." https://omanscience.com/en/articles/speaking-the-navigator-s-language-trajectory-grounded-instruction-translation-for-frozen-aerial-vln-agents.
Harvard
Chen, X., Liu, Z., Xu, X., Xu, J., Zhou, C., Su, Y., Zeng, R., Du, T., Yao, K., Li, C. and Ji, S. (2026) 'Speaking the Navigator's Language: Trajectory-Grounded Instruction Translation for Frozen Aerial VLN Agents', Available at: https://omanscience.com/en/articles/speaking-the-navigator-s-language-trajectory-grounded-instruction-translation-for-frozen-aerial-vln-agents.
Vancouver
Chen X, Liu Z, Xu X, Xu J, Zhou C, Su Y, et al. Speaking the Navigator's Language: Trajectory-Grounded Instruction Translation for Frozen Aerial VLN Agents. https://omanscience.com/en/articles/speaking-the-navigator-s-language-trajectory-grounded-instruction-translation-for-frozen-aerial-vln-agents
IEEE
X. Chen, Z. Liu, X. Xu, J. Xu, C. Zhou, Y. Su, R. Zeng, T. Du, K. Yao, C. Li, and S. Ji, "Speaking the Navigator's Language: Trajectory-Grounded Instruction Translation for Frozen Aerial VLN Agents," https://omanscience.com/en/articles/speaking-the-navigator-s-language-trajectory-grounded-instruction-translation-for-frozen-aerial-vln-agents.