Research map / Label-efficient and robust learning / Recognition and long-tail
ABSent: Cross-Lingual Sentence Representation Mapping with Bidirectional GANs
Zuohui Fu, Yikun Xian, Shijie Geng, Yingqiang Ge, Yuting Wang, Xin Dong, Guang Wang, Gerard de Melo
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 bidirectional mapping rather than unidirectional mapping (ablation) | ABSent · Tatoeba and Europarl, four language directions | precision@1 (%): 26.9 -> 34.4; 29.1 -> 34.9; 27.4 -> 31.8; 25.3 -> 30.0; 21.5 -> 24.7; 20.0 -> 23.1; 39.3 -> 45.0; 36.9 -> 45.8 | Improved |
| Add mismatch-pair loss (ablation) | ABSent and uni-Sent · Tatoeba and Europarl, four language directions | precision@1 (%): 32.0 -> 34.4; 32.9 -> 34.9; 30.7 -> 31.8; 29.6 -> 30.0; 23.9 -> 24.7; 22.8 -> 23.1; 43.0 -> 45.0; 44.5 -> 45.8 | Improved |
| Choose the corpus-appropriate sentence weighting (simple averaging for Tatoeba, TF-IDF for Europarl) (ablation) | ABSent · Tatoeba and Europarl, four language directions | precision@1 (%): 27.4 -> 34.4; 28.1 -> 34.9; 28.0 -> 31.8; 26.2 -> 30.0; 21.9 -> 24.7; 22.4 -> 23.1; 37.1 -> 45.0; 38.4 -> 45.8 | Mixed |
| Align individual words before forming sentence embeddings (ablation) | ABSent · Europarl, four language directions | precision@1 (%): 24.7 -> 21.9; 23.1 -> 22.4; 45.0 -> 37.1; 45.8 -> 38.4 | Hurt |
Closest papers by meaning
- Translation-Based Matching Adversarial Network for Cross-Lingual Natural Language Inference (AAAI 2020 · similarity 0.88)
- Unsupervised Interlingual Semantic Representations from Sentence Embeddings for Zero-Shot Cross-Lingual Transfer (AAAI 2020 · similarity 0.87)
- Comparing Unsupervised Word Translation Methods Step by Step (NeurIPS 2019 · similarity 0.86)
- Cross-lingual Alignment vs Joint Training: A Comparative Study and A Simple Unified Framework (ICLR 2020 · similarity 0.86)
- CL2CM: Improving Cross-Lingual Cross-Modal Retrieval via Cross-Lingual Knowledge Transfer (AAAI 2024 · similarity 0.84)
- Narrowing the Gap between Supervised and Unsupervised Sentence Representation Learning with Large Language Model (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.
