{"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/improving-reinforcement-learning-based-image","title":"Improving Reinforcement Learning Based Image Captioning with Natural Language Prior","arxiv_id":"1809.06227","date":"2018-09-13","proceeding":"EMNLP 2018 10","authors":["Tszhang Guo","Shiyu Chang","Mo Yu","Kun Bai"],"abstract":"Recently, Reinforcement Learning (RL) approaches have demonstrated advanced\nperformance in image captioning by directly optimizing the metric used for\ntesting. However, this shaped reward introduces learning biases, which reduces\nthe readability of generated text. In addition, the large sample space makes\ntraining unstable and slow. To alleviate these issues, we propose a simple\ncoherent solution that constrains the action space using an n-gram language\nprior. Quantitative and qualitative evaluations on benchmarks show that RL with\nthe simple add-on module performs favorably against its counterpart in terms of\nboth readability and speed of convergence. Human evaluation results show that\nour model is more human readable and graceful. The implementation will become\npublicly available upon the acceptance of the paper.","url_abs":"http://arxiv.org/abs/1809.06227v1","url_pdf":"http://arxiv.org/pdf/1809.06227v1.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":"improving-reinforcement-learning-based-image","repo_url":"https://github.com/tgGuo15/PriorImageCaption","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"image-captioning","task_name":"Image Captioning"},{"task_slug":"reinforcement-learning","task_name":"Reinforcement Learning"},{"task_slug":"reinforcement-learning-1","task_name":"Reinforcement Learning (RL)"},{"task_slug":"reinforcement-learning-2","task_name":"reinforcement-learning"}],"methods":[{"method_slug":"speed","method_name":"SPEED"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}