Research map / Label-efficient and robust learning / Recognition and long-tail
FRAGE: Frequency-Agnostic Word Representation
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 |
|---|---|---|---|
| add adversarial frequency discriminator on embeddings (FRAGE) | Skip-gram · RW | Spearman: 52.67 -> 58.12 | Helped |
| add adversarial frequency discriminator on embeddings (FRAGE) | Skip-gram · RG65 | Spearman: 75.63 -> 78.78 | Helped |
| add adversarial frequency discriminator on embeddings (FRAGE) | AWD-LSTM · PTB test | perplexity: 57.3 -> 56.1 | Helped |
| add adversarial frequency discriminator on embeddings (FRAGE) | AWD-LSTM-MoS · PTB test | perplexity: 54.44 -> 53.31 | Helped |
| add adversarial frequency discriminator on embeddings (FRAGE) | AWD-LSTM · WT2 test | perplexity: 65.8 -> 63.4 | Helped |
| add adversarial frequency discriminator on embeddings (FRAGE) | AWD-LSTM-MoS + dynamic evaluation · WT2 test | perplexity: 40.68 -> 39.14 | Helped |
| add adversarial frequency discriminator on embeddings (FRAGE) | Transformer Base · WMT14 En-De | BLEU: 27.30 -> 28.36 | Helped |
| add adversarial frequency discriminator on embeddings (FRAGE) | Transformer Big · WMT14 En-De | BLEU: 28.40 -> 29.11 | Helped |
| add adversarial frequency discriminator on embeddings (FRAGE) | Transformer · IWSLT14 De-En | BLEU: 33.12 -> 33.97 | Helped |
| add adversarial frequency discriminator on embeddings (FRAGE) | RCNN · AG's News | accuracy: 90.47% -> 91.73% | Helped |
| add adversarial frequency discriminator on embeddings (FRAGE) | RCNN · IMDB | accuracy: 92.41% -> 93.07% | Helped |
| add adversarial frequency discriminator on embeddings (FRAGE) | RCNN · 20 Newsgroups | accuracy: 96.49% -> 96.93% | Helped |
Closest papers by meaning
- Learning Context-Specific Word/Character Embeddings (AAAI 2017 · similarity 0.88)
- Zipfian Whitening (NeurIPS 2024 · similarity 0.87)
- Learning Semantic Representations for Novel Words: Leveraging Both Form and Context (AAAI 2019 · similarity 0.87)
- Understanding the Semantic Content of Sparse Word Embeddings Using a Commonsense Knowledge Base (AAAI 2020 · similarity 0.87)
- Closed-Form Training Dynamics Reveal Learned Features and Linear Structure in Word2Vec-like Models (NeurIPS 2025 · similarity 0.85)
- By Tying Embeddings You Are Assuming the Distributional Hypothesis (ICML 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.