Papers › FRAGE: Frequency-Agnostic Word Representation
FRAGE: Frequency-Agnostic Word Representation
Chengyue Gong, Di He, Xu Tan, Tao Qin, Li-Wei Wang, Tie-Yan Liu
Continuous word representation (aka word embedding) is a basic building block in many neural network-based models used in natural language processing tasks. Although it is widely accepted that words with similar semantics should be close to each other in the embedding space, we find that word embeddings learned in several tasks are biased towards word frequency: the embeddings of high-frequency and low-frequency words lie in different subregions of the embedding space, and the embedding of a rare word and a popular word can be far from each other even if they are semantically similar. This makes learned word embeddings ineffective, especially for rare words, and consequently limits the performance of these neural network models. In this paper, we develop a neat, simple yet effective way to learn \emph{FRequency-AGnostic word Embedding} (FRAGE) using adversarial training. We conducted comprehensive studies on ten datasets across four natural language processing tasks, including word similarity, language modeling, machine translation and text classification. Results show that with FRAGE, we achieve higher performance than the baselines in all tasks.
In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.
Code
Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.
Code Syntology ran Syntology
Not run by Syntology. Nothing on this page verifies that the listed code works.
Tasks
Results from the paper archive 2025-07-28
| Task | Dataset | Model | Metric | Value | Rank at snapshot | Leaderboard | Report |
|---|---|---|---|---|---|---|---|
| Language Modelling | Penn Treebank (Word Level) | FRAGE + AWD-LSTM-MoS + dynamic eval | Params | 22M | #7 of 43 | Archive leaderboard | report |
| Language Modelling | Penn Treebank (Word Level) | FRAGE + AWD-LSTM-MoS + dynamic eval | Test perplexity | 46.54 | #7 of 43 | Archive leaderboard | report |
| Language Modelling | Penn Treebank (Word Level) | FRAGE + AWD-LSTM-MoS + dynamic eval | Validation perplexity | 47.38 | #7 of 43 | Archive leaderboard | report |
| Language Modelling | WikiText-2 | FRAGE + AWD-LSTM-MoS + dynamic eval | Number of params | 35M | #13 of 38 | Archive leaderboard | report |
| Language Modelling | WikiText-2 | FRAGE + AWD-LSTM-MoS + dynamic eval | Test perplexity | 39.14 | #13 of 38 | Archive leaderboard | report |
| Language Modelling | WikiText-2 | FRAGE + AWD-LSTM-MoS + dynamic eval | Validation perplexity | 40.85 | #13 of 38 | Archive leaderboard | report |
| Machine Translation | IWSLT2015 German-English | Transformer with FRAGE | BLEU score | 33.97 | #3 of 15 | Archive leaderboard | report |
| Machine Translation | WMT2014 English-German | Transformer Big with FRAGE | BLEU score | 29.11 | #31 of 91 | Archive leaderboard | report |
Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.
Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections