Papers › MvP: Multi-view Prompting Improves Aspect Sentiment Tuple Prediction

MvP: Multi-view Prompting Improves Aspect Sentiment Tuple Prediction

22 May 2023arXiv:2305.12627archive 2025-07-28

Zhibin Gou, Qingyan Guo, Yujiu Yang

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.

PaperPDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

ZubinGou/multi-view-prompting officialmentioned in papermentioned on GitHubpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Aspect Category DetectionAspect Category PolarityAspect Sentiment Triplet ExtractionAspect-Based Sentiment AnalysisAspect-Based Sentiment Analysis (ABSA)Aspect-Category-Opinion-Sentiment Quadruple ExtractionAspect-Sentiment-Opinion Triplet ExtractionDiversityExtract AspectExtract aspect-polarity tupleHidden Aspect DetectionLanguage ModelingLanguage ModellingSentiment AnalysisStructured PredictionTerm Extraction

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Aspect-Based Sentiment Analysis (ABSA) ACOS MvP F1 (Laptop) 43.92 #1 of 9 Archive leaderboard report
Aspect-Based Sentiment Analysis (ABSA) ACOS MvP F1 (Restaurant) 61.54 #1 of 9 Archive leaderboard report
Aspect-Based Sentiment Analysis (ABSA) ACOS MvP (muilti-task) F1 (Laptop) 43.84 #2 of 9 Archive leaderboard report
Aspect-Based Sentiment Analysis (ABSA) ACOS MvP (muilti-task) F1 (Restaurant) 60.36 #2 of 9 Archive leaderboard report
Aspect-Based Sentiment Analysis (ABSA) ACOS ChatGPT (gpt-3.5-turbo, few-shot) F1 (Restaurant) 37.71 #8 of 9 Archive leaderboard report
Aspect-Based Sentiment Analysis (ABSA) ACOS ChatGPT (gpt-3.5-turbo, zero-shot) F1 (Restaurant) 27.11 #9 of 9 Archive leaderboard report
Aspect-Based Sentiment Analysis (ABSA) ASQP MvP (multi-task) F1 (R15) 52.21 #1 of 12 Archive leaderboard report
Aspect-Based Sentiment Analysis (ABSA) ASQP MvP (multi-task) F1 (R16) 58.94 #1 of 12 Archive leaderboard report
Aspect-Based Sentiment Analysis (ABSA) ASQP MvP F1 (R15) 51.04 #2 of 12 Archive leaderboard report
Aspect-Based Sentiment Analysis (ABSA) ASQP MvP F1 (R16) 60.39 #2 of 12 Archive leaderboard report
Aspect-Based Sentiment Analysis (ABSA) ASQP ChatGPT (gpt-3.5-turbo, few-shot) F1 (R15) 34.27 #11 of 12 Archive leaderboard report
Aspect-Based Sentiment Analysis (ABSA) ASQP ChatGPT (gpt-3.5-turbo, zero-shot) F1 (R15) 22.87 #12 of 12 Archive leaderboard report
Aspect-Based Sentiment Analysis (ABSA) ASTE MvP (multi-task) F1 (L14) 65.30 #1 of 13 Archive leaderboard report
Aspect-Based Sentiment Analysis (ABSA) ASTE MvP (multi-task) F1 (R15) 69.44 #1 of 13 Archive leaderboard report
Aspect-Based Sentiment Analysis (ABSA) ASTE MvP (multi-task) F1 (R16) 73.10 #1 of 13 Archive leaderboard report
Aspect-Based Sentiment Analysis (ABSA) ASTE MvP (multi-task) F1(R14) 76.30 #1 of 13 Archive leaderboard report
Aspect-Based Sentiment Analysis (ABSA) ASTE MvP F1 (L14) 63.33 #4 of 13 Archive leaderboard report
Aspect-Based Sentiment Analysis (ABSA) ASTE MvP F1 (R15) 65.89 #4 of 13 Archive leaderboard report
Aspect-Based Sentiment Analysis (ABSA) ASTE MvP F1 (R16) 73.48 #4 of 13 Archive leaderboard report
Aspect-Based Sentiment Analysis (ABSA) ASTE MvP F1(R14) 74.05 #4 of 13 Archive leaderboard report
Aspect-Based Sentiment Analysis (ABSA) ASTE ChatGPT (gpt-3.5-turbo, few-shot) F1 (L14) 38.12 #12 of 13 Archive leaderboard report
Aspect-Based Sentiment Analysis (ABSA) ASTE ChatGPT (gpt-3.5-turbo, zero-shot) F1 (L14) 36.05 #13 of 13 Archive leaderboard report
Aspect-Based Sentiment Analysis (ABSA) TASD MvP (multi-task) F1 (R15) 64.74 #1 of 11 Archive leaderboard report
Aspect-Based Sentiment Analysis (ABSA) TASD MvP (multi-task) F1 (R16) 70.18 #1 of 11 Archive leaderboard report
Aspect-Based Sentiment Analysis (ABSA) TASD MvP F1 (R15) 64.53 #2 of 11 Archive leaderboard report
Aspect-Based Sentiment Analysis (ABSA) TASD MvP F1 (R16) 72.76 #2 of 11 Archive leaderboard report
Aspect-Based Sentiment Analysis (ABSA) TASD ChatGPT (gpt-3.5-turbo, few-shot) F1 (R16) 46.51 #10 of 11 Archive leaderboard report
Aspect-Based Sentiment Analysis (ABSA) TASD ChatGPT (gpt-3.5-turbo, zero-shot) F1 (R16) 34.08 #11 of 11 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections