{"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/quantifying-the-amount-of-visual-information","title":"Quantifying the amount of visual information used by neural caption generators","arxiv_id":"1810.05475","date":"2018-10-12","proceeding":null,"authors":["Marc Tanti","Albert Gatt","Kenneth P. Camilleri"],"abstract":"This paper addresses the sensitivity of neural image caption generators to\ntheir visual input. A sensitivity analysis and omission analysis based on image\nfoils is reported, showing that the extent to which image captioning\narchitectures retain and are sensitive to visual information varies depending\non the type of word being generated and the position in the caption as a whole.\nWe motivate this work in the context of broader goals in the field to achieve\nmore explainability in AI.","url_abs":"http://arxiv.org/abs/1810.05475v1","url_pdf":"http://arxiv.org/pdf/1810.05475v1.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":"quantifying-the-amount-of-visual-information","repo_url":"https://github.com/mtanti/quantifing-visual-information","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"tf","reach":null}],"tasks":[{"task_slug":"image-captioning","task_name":"Image Captioning"},{"task_slug":null,"task_name":"Position"},{"task_slug":"sensitivity","task_name":"Sensitivity"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}