{"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/are-you-talking-to-a-machine-dataset-and","title":"Are You Talking to a Machine? Dataset and Methods for Multilingual Image Question Answering","arxiv_id":"1505.05612","date":"2015-05-21","proceeding":"NeurIPS 2015","authors":["Haoyuan Gao","Junhua Mao","Jie zhou","Zhiheng Huang","Lei Wang","Wei Xu"],"abstract":"In this paper, we present the mQA model, which is able to answer questions\nabout the content of an image. The answer can be a sentence, a phrase or a\nsingle word. Our model contains four components: a Long Short-Term Memory\n(LSTM) to extract the question representation, a Convolutional Neural Network\n(CNN) to extract the visual representation, an LSTM for storing the linguistic\ncontext in an answer, and a fusing component to combine the information from\nthe first three components and generate the answer. We construct a Freestyle\nMultilingual Image Question Answering (FM-IQA) dataset to train and evaluate\nour mQA model. It contains over 150,000 images and 310,000 freestyle Chinese\nquestion-answer pairs and their English translations. The quality of the\ngenerated answers of our mQA model on this dataset is evaluated by human judges\nthrough a Turing Test. Specifically, we mix the answers provided by humans and\nour model. The human judges need to distinguish our model from the human. They\nwill also provide a score (i.e. 0, 1, 2, the larger the better) indicating the\nquality of the answer. We propose strategies to monitor the quality of this\nevaluation process. The experiments show that in 64.7% of cases, the human\njudges cannot distinguish our model from humans. The average score is 1.454\n(1.918 for human). The details of this work, including the FM-IQA dataset, can\nbe found on the project page: http://idl.baidu.com/FM-IQA.html","url_abs":"http://arxiv.org/abs/1505.05612v3","url_pdf":"http://arxiv.org/pdf/1505.05612v3.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":"are-you-talking-to-a-machine-dataset-and","repo_url":"https://github.com/miohana/vqa","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"question-answering","task_name":"Question Answering"},{"task_slug":"sentence","task_name":"Sentence"},{"task_slug":"visual-question-answering","task_name":"Visual Question Answering (VQA)"}],"methods":[{"method_slug":"lstm","method_name":"LSTM"},{"method_slug":"sigmoid-activation","method_name":"Sigmoid Activation"},{"method_slug":"tanh-activation","method_name":"Tanh Activation"}],"datasets_introduced":[{"slug":"fm-iqa","name":"FM-IQA","full_name":"Freestyle Multilingual Image Question Answering"}],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1505.05612","atlas_url":"https://app.syntology.ai/?focus=1505.05612","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}