{"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/lego-absa-a-prompt-based-task-assemblable","title":"LEGO-ABSA: A Prompt-based Task Assemblable Unified Generative Framework for Multi-task Aspect-based Sentiment Analysis","arxiv_id":null,"date":"2022-10-01","proceeding":"COLING 2022 10","authors":["Tianhao Gao","Jun Fang","Hanyu Liu","Zhiyuan Liu","Chao Liu","Pengzhang Liu","Yongjun Bao","Weipeng Yan"],"abstract":"Aspect-based sentiment analysis (ABSA) has received increasing attention recently. ABSA can be divided into multiple tasks according to the different extracted elements. Existing generative methods usually treat the output as a whole string rather than the combination of different elements and only focus on a single task at once. This paper proposes a unified generative multi-task framework that can solve multiple ABSA tasks by controlling the type of task prompts consisting of multiple element prompts. Further, the proposed approach can train on simple tasks and transfer to difficult tasks by assembling task prompts, like assembling Lego bricks. We conduct experiments on six ABSA tasks across multiple benchmarks. Our proposed multi-task approach achieves new state-of-the-art results in almost all tasks and competitive results in task transfer scenarios.","url_abs":"https://aclanthology.org/2022.coling-1.610","url_pdf":"https://aclanthology.org/2022.coling-1.610.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":[],"tasks":[{"task_slug":"aspect-based-sentiment-analysis-1","task_name":"Aspect-Based Sentiment Analysis"},{"task_slug":"aspect-based-sentiment-analysis","task_name":"Aspect-Based Sentiment Analysis (ABSA)"},{"task_slug":"sentiment-analysis","task_name":"Sentiment Analysis"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/aspect-based-sentiment-analysis-absa-on-asqp","task":"Aspect-Based Sentiment Analysis (ABSA)","dataset":"ASQP","model":"LEGO-ABSA (multi-task)","rank_in_archive_order":6,"of":12,"metrics":{"F1 (R15)":"46.10","F1 (R16)":"57.60"},"uses_additional_data":true},{"leaderboard":"/sota/aspect-based-sentiment-analysis-absa-on-aste","task":"Aspect-Based Sentiment Analysis (ABSA)","dataset":"ASTE","model":"LEGO-ABSA (multi-task)","rank_in_archive_order":7,"of":13,"metrics":{"F1 (L14)":"62.20","F1 (R15)":"64.40","F1 (R16)":"69.90","F1(R14)":"73.70"},"uses_additional_data":true},{"leaderboard":"/sota/aspect-based-sentiment-analysis-absa-on-tasd","task":"Aspect-Based Sentiment Analysis (ABSA)","dataset":"TASD","model":"LEGO-ABSA (multi-task)","rank_in_archive_order":5,"of":11,"metrics":{"F1 (R15)":"62.30","F1 (R16)":"71.80"},"uses_additional_data":true}],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}