{"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-task-attentive-residual-networks-for","title":"Multi-Task Attentive Residual Networks for Argument Mining","arxiv_id":"2102.12227","date":"2021-02-24","proceeding":null,"authors":["Andrea Galassi","Marco Lippi","Paolo Torroni"],"abstract":"We explore the use of residual networks and neural attention for multiple argument mining tasks. We propose a residual architecture that exploits attention, multi-task learning, and makes use of ensemble, without any assumption on document or argument structure. We present an extensive experimental evaluation on five different corpora of user-generated comments, scientific publications, and persuasive essays. Our results show that our approach is a strong competitor against state-of-the-art architectures with a higher computational footprint or corpus-specific design, representing an interesting compromise between generality, performance accuracy and reduced model size.","url_abs":"https://arxiv.org/abs/2102.12227v3","url_pdf":"https://arxiv.org/pdf/2102.12227v3.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-task-attentive-residual-networks-for","repo_url":"https://github.com/AGalassi/StructurePrediction18","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"argument-mining","task_name":"Argument Mining"},{"task_slug":"component-classification","task_name":"Component Classification"},{"task_slug":"link-prediction","task_name":"Link Prediction"},{"task_slug":"multi-task-learning","task_name":"Multi-Task Learning"},{"task_slug":"relation-classification","task_name":"Relation Classification"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/component-classification-on-cdcp","task":"Component Classification","dataset":"CDCP","model":"ResAttArg","rank_in_archive_order":1,"of":1,"metrics":{"Macro F1":"78.71"},"uses_additional_data":false},{"leaderboard":"/sota/link-prediction-on-abstrct-neoplasm","task":"Link Prediction","dataset":"AbstRCT - Neoplasm","model":"ResAttArg","rank_in_archive_order":1,"of":1,"metrics":{"F1":"54.43"},"uses_additional_data":false},{"leaderboard":"/sota/link-prediction-on-cdcp","task":"Link Prediction","dataset":"CDCP","model":"ResAttArg","rank_in_archive_order":1,"of":1,"metrics":{"F1":"29.73"},"uses_additional_data":false},{"leaderboard":"/sota/link-prediction-on-dr-inventor","task":"Link Prediction","dataset":"DRI Corpus","model":"ResAttArg","rank_in_archive_order":1,"of":1,"metrics":{"F1":"43.66"},"uses_additional_data":false},{"leaderboard":"/sota/relation-classification-on-abstrct-neoplasm","task":"Relation Classification","dataset":"AbstRCT - Neoplasm","model":"ResAttArg","rank_in_archive_order":1,"of":1,"metrics":{"Macro F1":"70.92"},"uses_additional_data":false},{"leaderboard":"/sota/relation-classification-on-cdcp","task":"Relation Classification","dataset":"CDCP","model":"ResAttArg","rank_in_archive_order":1,"of":1,"metrics":{"Macro F1":"42.95"},"uses_additional_data":false},{"leaderboard":"/sota/relation-classification-on-dr-inventor","task":"Relation Classification","dataset":"DRI Corpus","model":"ResAttArg","rank_in_archive_order":1,"of":1,"metrics":{"Macro F1":"37.72"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2102.12227","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}