{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/attention-based-natural-language-grounding-by","title":"Attention Based Natural Language Grounding by Navigating Virtual Environment","arxiv_id":"1804.08454","date":"2018-04-23","proceeding":null,"authors":["Akilesh B","Abhishek Sinha","Mausoom Sarkar","Balaji Krishnamurthy"],"abstract":"In this work, we focus on the problem of grounding language by training an\nagent to follow a set of natural language instructions and navigate to a target\nobject in an environment. The agent receives visual information through raw\npixels and a natural language instruction telling what task needs to be\nachieved and is trained in an end-to-end way. We develop an attention mechanism\nfor multi-modal fusion of visual and textual modalities that allows the agent\nto learn to complete the task and achieve language grounding. Our experimental\nresults show that our attention mechanism outperforms the existing multi-modal\nfusion mechanisms proposed for both 2D and 3D environments in order to solve\nthe above-mentioned task in terms of both speed and success rate. We show that\nthe learnt textual representations are semantically meaningful as they follow\nvector arithmetic in the embedding space. The effectiveness of our attention\napproach over the contemporary fusion mechanisms is also highlighted from the\ntextual embeddings learnt by the different approaches. We also show that our\nmodel generalizes effectively to unseen scenarios and exhibit zero-shot\ngeneralization capabilities both in 2D and 3D environments. The code for our 2D\nenvironment as well as the models that we developed for both 2D and 3D are\navailable at https://github.com/rl-lang-grounding/rl-lang-ground.","url_abs":"http://arxiv.org/abs/1804.08454v2","url_pdf":"http://arxiv.org/pdf/1804.08454v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"attention-based-natural-language-grounding-by","repo_url":"https://github.com/rl-lang-grounding/rl-lang-ground","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"tf","reach":null}],"tasks":[{"task_slug":"navigate","task_name":"Navigate"},{"task_slug":"zero-shot-generalization","task_name":"Zero-shot Generalization"}],"methods":[{"method_slug":"speed","method_name":"SPEED"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}