{"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/multi-attention-recurrent-network-for-human","title":"Multi-attention Recurrent Network for Human Communication Comprehension","arxiv_id":"1802.00923","date":"2018-02-03","proceeding":null,"authors":["Amir Zadeh","Paul Pu Liang","Soujanya Poria","Prateek Vij","Erik Cambria","Louis-Philippe Morency"],"abstract":"Human face-to-face communication is a complex multimodal signal. We use words\n(language modality), gestures (vision modality) and changes in tone (acoustic\nmodality) to convey our intentions. Humans easily process and understand\nface-to-face communication, however, comprehending this form of communication\nremains a significant challenge for Artificial Intelligence (AI). AI must\nunderstand each modality and the interactions between them that shape human\ncommunication. In this paper, we present a novel neural architecture for\nunderstanding human communication called the Multi-attention Recurrent Network\n(MARN). The main strength of our model comes from discovering interactions\nbetween modalities through time using a neural component called the\nMulti-attention Block (MAB) and storing them in the hybrid memory of a\nrecurrent component called the Long-short Term Hybrid Memory (LSTHM). We\nperform extensive comparisons on six publicly available datasets for multimodal\nsentiment analysis, speaker trait recognition and emotion recognition. MARN\nshows state-of-the-art performance on all the datasets.","url_abs":"http://arxiv.org/abs/1802.00923v1","url_pdf":"http://arxiv.org/pdf/1802.00923v1.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":"multi-attention-recurrent-network-for-human","repo_url":"https://github.com/clin366/MFN","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"multi-attention-recurrent-network-for-human","repo_url":"https://github.com/pliang279/MFN","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"emotion-recognition","task_name":"Emotion Recognition"},{"task_slug":"multimodal-sentiment-analysis","task_name":"Multimodal Sentiment Analysis"},{"task_slug":"sentiment-analysis","task_name":"Sentiment Analysis"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/multimodal-sentiment-analysis-on-mosi","task":"Multimodal Sentiment Analysis","dataset":"MOSI","model":"MARN","rank_in_archive_order":10,"of":11,"metrics":{"Accuracy":"77.1%"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1802.00923","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}