Research map / Label-efficient and robust learning / Recognition and long-tail
Learning to Attend via Word-Aspect Associative Fusion for Aspect-Based Sentiment Analysis
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 |
|---|---|---|---|
| Use circular convolution for word-aspect fusion rather than circular correlation (ablation) | AF-LSTM · ABSA benchmarks | accuracy: 68.81 vs 64.89; 83.58 vs 79.96; 75.44 vs 74.76; 87.78 vs 86.91; 81.29 vs 80.47; 87.26 vs 86.58; 78.44 vs 74.68; 81.49 vs | Helped |
| Replace circular convolution with elementwise multiplication for fusion (ablation) | AF-LSTM · ABSA benchmarks | accuracy: 68.81 vs 66.14; 83.58 vs 83.37; 75.44 vs 75.35; 87.78 vs 86.47; 81.29 vs 79.96; 87.26 vs 81.71; 78.44 vs 77.44; 81.49 vs | Hurt |
| Replace aspect-word concatenation at the LSTM input with the AF-LSTM fusion approach | AF-LSTM (CONV) versus ATAE-LSTM · ABSA benchmark settings | accuracy | Helped |
| Add circular convolution-based word-aspect fusion rather than attention-layer aspect concatenation | AF-LSTM (CONV) versus AT-LSTM · ABSA benchmarks | accuracy | Helped |
| Add the optional projection combining attentive representation with the last hidden state (ablation) | AF-LSTM · ABSA | accuracy | No clear effect |
| Add normalization before associating hidden states and aspect vectors (ablation) | AF-LSTM · ABSA | accuracy | No clear effect |
| Use batch normalization as the association-layer normalization (ablation) | AF-LSTM · Some datasets | accuracy | Helped |
| Remove aspect information from attention (ablation) | AT-LSTM (w/o aspect) · Restaurants contrasting-polarity example | attention focus correctness | Hurt |
| Add aspect-aware word-aspect fusion instead of aspect-free attention | AF-LSTM (CONV) · Restaurants contrasting-polarity example | attention focus correctness | 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)
- You Only Read Once: Constituency-Oriented Relational Graph Convolutional Network for Multi-Aspect Multi-Sentiment Classification (AAAI 2024 · similarity 0.89)
- Exploring Model Editing for LLM-based Aspect-Based Sentiment Classification (AAAI 2025 · similarity 0.88)
- Aspect-Based Sentiment Analysis with Explicit Sentiment Augmentations (AAAI 2024 · similarity 0.88)
- TextGT: A Double-View Graph Transformer on Text for Aspect-Based Sentiment Analysis (AAAI 2024 · similarity 0.87)
- A Novel Energy Based Model Mechanism for Multi-Modal Aspect-Based Sentiment Analysis (AAAI 2024 · similarity 0.86)
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.