{"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/hadamard-product-for-low-rank-bilinear","title":"Hadamard Product for Low-rank Bilinear Pooling","arxiv_id":"1610.04325","date":"2016-10-14","proceeding":null,"authors":["Jin-Hwa Kim","Kyoung-Woon On","Woosang Lim","Jeonghee Kim","Jung-Woo Ha","Byoung-Tak Zhang"],"abstract":"Bilinear models provide rich representations compared with linear models.\nThey have been applied in various visual tasks, such as object recognition,\nsegmentation, and visual question-answering, to get state-of-the-art\nperformances taking advantage of the expanded representations. However,\nbilinear representations tend to be high-dimensional, limiting the\napplicability to computationally complex tasks. We propose low-rank bilinear\npooling using Hadamard product for an efficient attention mechanism of\nmultimodal learning. We show that our model outperforms compact bilinear\npooling in visual question-answering tasks with the state-of-the-art results on\nthe VQA dataset, having a better parsimonious property.","url_abs":"http://arxiv.org/abs/1610.04325v4","url_pdf":"http://arxiv.org/pdf/1610.04325v4.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":"hadamard-product-for-low-rank-bilinear","repo_url":"https://github.com/jnhwkim/MulLowBiVQA","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"torch","reach":{"status":"ok","spdx":"NOASSERTION"}},{"paper_slug":"hadamard-product-for-low-rank-bilinear","repo_url":"https://github.com/Adam1679/mutan-article-net","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"hadamard-product-for-low-rank-bilinear","repo_url":"https://github.com/Cadene/vqa.pytorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"hadamard-product-for-low-rank-bilinear","repo_url":"https://github.com/facebookresearch/ParlAI","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"hadamard-product-for-low-rank-bilinear","repo_url":"https://github.com/gabegrand/adversarial-vqa","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"hadamard-product-for-low-rank-bilinear","repo_url":"https://github.com/jnhwkim/nips-mrn-vqa","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"torch","reach":{"status":"ok","spdx":"NOASSERTION"}},{"paper_slug":"hadamard-product-for-low-rank-bilinear","repo_url":"https://github.com/vuhoangminh/vqa_medical","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"hadamard-product-for-low-rank-bilinear","repo_url":"https://github.com/yikang-li/iqan","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"hadamard-product-for-low-rank-bilinear","repo_url":"https://github.com/MindSpore-scientific-2/code-10/tree/main/PR_Product","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null},{"paper_slug":"hadamard-product-for-low-rank-bilinear","repo_url":"https://github.com/MindSpore-scientific-2/code-8/tree/main/PR_Product","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null},{"paper_slug":"hadamard-product-for-low-rank-bilinear","repo_url":"https://github.com/MindSpore-scientific/code-8/tree/main/PR_Product","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null}],"tasks":[{"task_slug":"visual-question-answering-1","task_name":"Visual Question Answering"},{"task_slug":"visual-question-answering","task_name":"Visual Question Answering (VQA)"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1610.04325","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}