Papers › Improving Neural Language Modeling via Adversarial Training
Improving Neural Language Modeling via Adversarial Training
Dilin Wang, Chengyue Gong, Qiang Liu
Recently, substantial progress has been made in language modeling by using deep neural networks. However, in practice, large scale neural language models have been shown to be prone to overfitting. In this paper, we present a simple yet highly effective adversarial training mechanism for regularizing neural language models. The idea is to introduce adversarial noise to the output embedding layer while training the models. We show that the optimal adversarial noise yields a simple closed-form solution, thus allowing us to develop a simple and time efficient algorithm. Theoretically, we show that our adversarial mechanism effectively encourages the diversity of the embedding vectors, helping to increase the robustness of models. Empirically, we show that our method improves on the single model state-of-the-art results for language modeling on Penn Treebank (PTB) and Wikitext-2, achieving test perplexity scores of 46.01 and 38.07, respectively. When applied to machine translation, our method improves over various transformer-based translation baselines in BLEU scores on the WMT14 English-German and IWSLT14 German-English 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) | adversarial + AWD-LSTM-MoS + dynamic eval | Params | 22M | #5 of 43 | Archive leaderboard | report |
| Language Modelling | Penn Treebank (Word Level) | adversarial + AWD-LSTM-MoS + dynamic eval | Test perplexity | 46.01 | #5 of 43 | Archive leaderboard | report |
| Language Modelling | Penn Treebank (Word Level) | adversarial + AWD-LSTM-MoS + dynamic eval | Validation perplexity | 46.63 | #5 of 43 | Archive leaderboard | report |
| Language Modelling | WikiText-103 | AdvSoft (+ 4 layer QRNN + dynamic eval) | Test perplexity | 28.0 | #65 of 89 | Archive leaderboard | report |
| Language Modelling | WikiText-103 | AdvSoft (+ 4 layer QRNN + dynamic eval) | Validation perplexity | 27.2 | #65 of 89 | Archive leaderboard | report |
| Language Modelling | WikiText-2 | adversarial + AWD-LSTM-MoS + dynamic eval | Number of params | 35M | #12 of 38 | Archive leaderboard | report |
| Language Modelling | WikiText-2 | adversarial + AWD-LSTM-MoS + dynamic eval | Test perplexity | 38.65 | #12 of 38 | Archive leaderboard | report |
| Language Modelling | WikiText-2 | adversarial + AWD-LSTM-MoS + dynamic eval | Validation perplexity | 40.27 | #12 of 38 | Archive leaderboard | report |
| Machine Translation | WMT2014 English-German | Transformer Big + adversarial MLE | BLEU score | 29.52 | #24 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