{"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/mvp-multi-view-prompting-improves-aspect","title":"MvP: Multi-view Prompting Improves Aspect Sentiment Tuple Prediction","arxiv_id":"2305.12627","date":"2023-05-22","proceeding":null,"authors":["Zhibin Gou","Qingyan Guo","Yujiu Yang"],"abstract":"Generative methods greatly promote aspect-based sentiment analysis via generating a sequence of sentiment elements in a specified format. However, existing studies usually predict sentiment elements in a fixed order, which ignores the effect of the interdependence of the elements in a sentiment tuple and the diversity of language expression on the results. In this work, we propose Multi-view Prompting (MvP) that aggregates sentiment elements generated in different orders, leveraging the intuition of human-like problem-solving processes from different views. Specifically, MvP introduces element order prompts to guide the language model to generate multiple sentiment tuples, each with a different element order, and then selects the most reasonable tuples by voting. MvP can naturally model multi-view and multi-task as permutations and combinations of elements, respectively, outperforming previous task-specific designed methods on multiple ABSA tasks with a single model. Extensive experiments show that MvP significantly advances the state-of-the-art performance on 10 datasets of 4 benchmark tasks, and performs quite effectively in low-resource settings. Detailed evaluation verified the effectiveness, flexibility, and cross-task transferability of MvP.","url_abs":"https://arxiv.org/abs/2305.12627v1","url_pdf":"https://arxiv.org/pdf/2305.12627v1.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":"mvp-multi-view-prompting-improves-aspect","repo_url":"https://github.com/ZubinGou/multi-view-prompting","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"aspect-category-detection","task_name":"Aspect Category Detection"},{"task_slug":"aspect-category-polarity","task_name":"Aspect Category Polarity"},{"task_slug":"aspect-sentiment-triplet-extraction","task_name":"Aspect Sentiment Triplet Extraction"},{"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":"aspect-category-opinion-sentiment-quadruple","task_name":"Aspect-Category-Opinion-Sentiment Quadruple Extraction"},{"task_slug":"aspect-sentiment-opinion-triplet-extraction","task_name":"Aspect-Sentiment-Opinion Triplet Extraction"},{"task_slug":"diversity","task_name":"Diversity"},{"task_slug":"extract-aspect","task_name":"Extract Aspect"},{"task_slug":"extract-aspect-polarity-tuple","task_name":"Extract aspect-polarity tuple"},{"task_slug":"hidden-aspect-detection","task_name":"Hidden Aspect Detection"},{"task_slug":"language-modeling","task_name":"Language Modeling"},{"task_slug":"language-modelling","task_name":"Language Modelling"},{"task_slug":"sentiment-analysis","task_name":"Sentiment Analysis"},{"task_slug":"structured-prediction","task_name":"Structured Prediction"},{"task_slug":"term-extraction","task_name":"Term Extraction"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/aspect-based-sentiment-analysis-absa-on-acos","task":"Aspect-Based Sentiment Analysis (ABSA)","dataset":"ACOS","model":"MvP","rank_in_archive_order":1,"of":9,"metrics":{"F1 (Laptop)":"43.92","F1 (Restaurant)":"61.54"},"uses_additional_data":false},{"leaderboard":"/sota/aspect-based-sentiment-analysis-absa-on-acos","task":"Aspect-Based Sentiment Analysis (ABSA)","dataset":"ACOS","model":"MvP (muilti-task)","rank_in_archive_order":2,"of":9,"metrics":{"F1 (Laptop)":"43.84","F1 (Restaurant)":"60.36"},"uses_additional_data":true},{"leaderboard":"/sota/aspect-based-sentiment-analysis-absa-on-acos","task":"Aspect-Based Sentiment Analysis (ABSA)","dataset":"ACOS","model":"ChatGPT (gpt-3.5-turbo, few-shot)","rank_in_archive_order":8,"of":9,"metrics":{"F1 (Restaurant)":"37.71"},"uses_additional_data":false},{"leaderboard":"/sota/aspect-based-sentiment-analysis-absa-on-acos","task":"Aspect-Based Sentiment Analysis (ABSA)","dataset":"ACOS","model":"ChatGPT (gpt-3.5-turbo, zero-shot)","rank_in_archive_order":9,"of":9,"metrics":{"F1 (Restaurant)":"27.11"},"uses_additional_data":false},{"leaderboard":"/sota/aspect-based-sentiment-analysis-absa-on-asqp","task":"Aspect-Based Sentiment Analysis (ABSA)","dataset":"ASQP","model":"MvP (multi-task)","rank_in_archive_order":1,"of":12,"metrics":{"F1 (R15)":"52.21","F1 (R16)":"58.94"},"uses_additional_data":true},{"leaderboard":"/sota/aspect-based-sentiment-analysis-absa-on-asqp","task":"Aspect-Based Sentiment Analysis (ABSA)","dataset":"ASQP","model":"MvP","rank_in_archive_order":2,"of":12,"metrics":{"F1 (R15)":"51.04","F1 (R16)":"60.39"},"uses_additional_data":false},{"leaderboard":"/sota/aspect-based-sentiment-analysis-absa-on-asqp","task":"Aspect-Based Sentiment Analysis (ABSA)","dataset":"ASQP","model":"ChatGPT (gpt-3.5-turbo, few-shot)","rank_in_archive_order":11,"of":12,"metrics":{"F1 (R15)":"34.27"},"uses_additional_data":false},{"leaderboard":"/sota/aspect-based-sentiment-analysis-absa-on-asqp","task":"Aspect-Based Sentiment Analysis (ABSA)","dataset":"ASQP","model":"ChatGPT (gpt-3.5-turbo, zero-shot)","rank_in_archive_order":12,"of":12,"metrics":{"F1 (R15)":"22.87"},"uses_additional_data":false},{"leaderboard":"/sota/aspect-based-sentiment-analysis-absa-on-aste","task":"Aspect-Based Sentiment Analysis (ABSA)","dataset":"ASTE","model":"MvP (multi-task)","rank_in_archive_order":1,"of":13,"metrics":{"F1 (L14)":"65.30","F1 (R15)":"69.44","F1 (R16)":"73.10","F1(R14)":"76.30"},"uses_additional_data":true},{"leaderboard":"/sota/aspect-based-sentiment-analysis-absa-on-aste","task":"Aspect-Based Sentiment Analysis (ABSA)","dataset":"ASTE","model":"MvP","rank_in_archive_order":4,"of":13,"metrics":{"F1 (L14)":"63.33","F1 (R15)":"65.89","F1 (R16)":"73.48","F1(R14)":"74.05"},"uses_additional_data":false},{"leaderboard":"/sota/aspect-based-sentiment-analysis-absa-on-aste","task":"Aspect-Based Sentiment Analysis (ABSA)","dataset":"ASTE","model":"ChatGPT (gpt-3.5-turbo, few-shot)","rank_in_archive_order":12,"of":13,"metrics":{"F1 (L14)":"38.12"},"uses_additional_data":false},{"leaderboard":"/sota/aspect-based-sentiment-analysis-absa-on-aste","task":"Aspect-Based Sentiment Analysis (ABSA)","dataset":"ASTE","model":"ChatGPT (gpt-3.5-turbo, zero-shot)","rank_in_archive_order":13,"of":13,"metrics":{"F1 (L14)":"36.05"},"uses_additional_data":false},{"leaderboard":"/sota/aspect-based-sentiment-analysis-absa-on-tasd","task":"Aspect-Based Sentiment Analysis (ABSA)","dataset":"TASD","model":"MvP (multi-task)","rank_in_archive_order":1,"of":11,"metrics":{"F1 (R15)":"64.74","F1 (R16)":"70.18"},"uses_additional_data":true},{"leaderboard":"/sota/aspect-based-sentiment-analysis-absa-on-tasd","task":"Aspect-Based Sentiment Analysis (ABSA)","dataset":"TASD","model":"MvP","rank_in_archive_order":2,"of":11,"metrics":{"F1 (R15)":"64.53","F1 (R16)":"72.76"},"uses_additional_data":false},{"leaderboard":"/sota/aspect-based-sentiment-analysis-absa-on-tasd","task":"Aspect-Based Sentiment Analysis (ABSA)","dataset":"TASD","model":"ChatGPT (gpt-3.5-turbo, few-shot)","rank_in_archive_order":10,"of":11,"metrics":{"F1 (R16)":"46.51"},"uses_additional_data":false},{"leaderboard":"/sota/aspect-based-sentiment-analysis-absa-on-tasd","task":"Aspect-Based Sentiment Analysis (ABSA)","dataset":"TASD","model":"ChatGPT (gpt-3.5-turbo, zero-shot)","rank_in_archive_order":11,"of":11,"metrics":{"F1 (R16)":"34.08"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2305.12627","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}