Research map / Label-efficient and robust learning / Domain adaptation and generalization
Simultaneous Learning of Pivots and Representations for Cross-Domain 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 |
|---|---|---|---|
| learn transferable pivots with the uncertainty-based Pivot Selector (800 to 500) (ablation) | TPT · Amazon product domain pairs (12 pairs) | avg accuracy (%): 82.3 -> 82.9 | Helped |
| use bidirectional Transformer with pivot masking instead of random mask (ablation) | Transformer · Amazon product domain pairs (12 pairs) | avg accuracy (%): 79.5 -> 82.3 | Helped |
| initialize with pretrained word2vec (ablation) | TPT · Amazon product domain pairs (12 pairs) | avg accuracy (%): 82.3 -> 82.1 | No clear effect |
| initialize with pretrained word2vec | CNN · Amazon product domain pairs (12 pairs) | avg accuracy (%): 71.6 -> 73.8 | Helped |
| increase fixed pivot number from 500 to 800 (ablation) | TPT w/o pivot learning · Amazon product domain pairs (12 pairs) | avg accuracy (%): 82.3 -> 82.2 | No clear effect |
| use TPT-selected transferable pivots in SCL (ablation) | SCL · Amazon product domain pairs (12 pairs) | avg accuracy (%): 74.3 -> 75.0 | Helped |
| replace PBLM recurrent LM with TPT | PBLM-CNN · Amazon product domain pairs (12 pairs) | avg accuracy (%): 80.4 -> 82.9 | Helped |
| TPT vs PBLM-LSTM on product to airline | TPT · Product to airline (4 pairs) | avg accuracy (%): 85.0 -> 85.6 | Helped |
Closest papers by meaning
- Structural Correspondence Learning for Cross-Lingual Sentiment Classification with One-to-Many Mappings (AAAI 2017 · similarity 0.88)
- Hierarchical Attention Transfer Network for Cross-Domain Sentiment Classification (AAAI 2018 · similarity 0.88)
- Interactive Attention Transfer Network for Cross-Domain Sentiment Classification (AAAI 2019 · similarity 0.86)
- RADAR: Relative Angular Divergence Across Representations (NeurIPS 2026 · similarity 0.82)
- Exploring Model Editing for LLM-based Aspect-Based Sentiment Classification (AAAI 2025 · similarity 0.82)
- You Only Read Once: Constituency-Oriented Relational Graph Convolutional Network for Multi-Aspect Multi-Sentiment Classification (AAAI 2024 · similarity 0.81)
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.