Research map / Label-efficient and robust learning / Recognition and long-tail
Exploiting Coarse-to-Fine Task Transfer for Aspect-Level Sentiment Classification
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| transfer coarse-grained AC knowledge using MGAN | AT-level sentiment classification · Laptop, Restaurant R2, Twitter | Accuracy: +4.13%, +3.58%, +5.26% | Helped |
| transfer coarse-grained AC knowledge using MGAN | AT-level sentiment classification · Laptop, Restaurant R2, Twitter | Macro-F1: +2.99%, +2.94%, +6.23% | Helped |
| add C2F attention to the source network to infer fine-grained aspect terms and position (ablation) | MGAN · Laptop, Restaurant R2, Twitter | Accuracy: +1.41%, +1.03%, +1.09% | Helped |
| add C2F attention to the source network to infer fine-grained aspect terms and position (ablation) | MGAN · Laptop, Restaurant R2, Twitter | Macro-F1: +1.79%, +3.62%, +1.16% | Helped |
| add position information to sentiment attention (ablation) | MGAN · Laptop, Restaurant R2, Twitter | Accuracy | Helped |
| add position information to sentiment attention (ablation) | MGAN · Laptop, Restaurant R2, Twitter | Macro-F1 | Helped |
| add semantic separation to contrastive feature alignment (ablation) | MGAN · All transfer pairs (average) | Accuracy: 0.81% | Helped |
| add semantic separation to contrastive feature alignment (ablation) | MGAN · All transfer pairs (average) | Macro-F1: 1.00% | Helped |
| use MGAN rather than M-DAN or M-MMD for transfer | AT-level sentiment classification · All transfer pairs (average) | Accuracy: 1.80%, 1.33% | Helped |
| use MGAN rather than M-DAN or M-MMD for transfer | AT-level sentiment classification · All transfer pairs (average) | Macro-F1: 1.90%, 1.66% | Helped |
Closest papers by meaning
- Modeling Adaptive Inter-Task Feature Interactions via Sentiment-Aware Contrastive Learning for Joint Aspect-Sentiment Prediction (AAAI 2024 · similarity 0.89)
- Exploring Model Editing for LLM-based Aspect-Based Sentiment Classification (AAAI 2025 · similarity 0.86)
- You Only Read Once: Constituency-Oriented Relational Graph Convolutional Network for Multi-Aspect Multi-Sentiment Classification (AAAI 2024 · similarity 0.86)
- TextGT: A Double-View Graph Transformer on Text for Aspect-Based Sentiment Analysis (AAAI 2024 · similarity 0.85)
- A Novel Energy Based Model Mechanism for Multi-Modal Aspect-Based Sentiment Analysis (AAAI 2024 · similarity 0.85)
- Aspect-Based Sentiment Analysis with Explicit Sentiment Augmentations (AAAI 2024 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.