Abstract

Embodied AI aims to develop robots that can understand and execute human language instructions, as well as communicate in natural languages. On this front, we study the task of generating highly detailed navigational instructions for the embodied robots to follow. Although recent studies have demonstrated significant leaps in the generation of step-by-step instructions from sequences of images, the generated instructions lack variety in terms of their referral to objects and landmarks. Existing speaker models learn strategies to evade the evaluation metrics and obtain higher scores even for low-quality sentences. In this work, we propose SAS (Spatially-Aware Speaker), an instruction generator or Speaker model that utilises both structural and semantic knowledge of the environment to produce richer instructions. For training, we employ a reward learning method in an adversarial setting to avoid systematic bias introduced by language evaluation metrics. Empirically, our method outperforms existing instruction generation models, evaluated using standard metrics. Our code is available at https://github.com/gmuraleekrishna/SAS.

Document Type

Conference Proceeding

Date of Publication

1-1-2024

Volume

1

School

Centre for Artificial Intelligence and Machine Learning (CAIML)

Creative Commons License

Creative Commons Attribution 4.0 License
This work is licensed under a Creative Commons Attribution 4.0 License.

Publisher

ACL Anthology

Comments

Gopinathan, M., Masek, M., Abu-Khalaf, J., & Suter, D. (2024). Spatially-aware speaker for vision-and-language navigation instruction generation. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) (pp. 13601–13614). Bangkok, Thailand: Association for Computational Linguistics. https://doi.org/10.18653/v1/2024.acl-long.734

Download

Share

 
COinS
 

Link to publisher version (DOI)

10.18653/v1/2024.acl-long.734