Papers › Mogrifier LSTM

Mogrifier LSTM

4 Sep 2019ICLR 2020 1arXiv:1909.01792archive 2025-07-28

Gábor Melis, Tomáš Kočiský, Phil Blunsom

Many advances in Natural Language Processing have been based upon more expressive models for how inputs interact with the context in which they occur. Recurrent networks, which have enjoyed a modicum of success, still lack the generalization and systematicity ultimately required for modelling language. In this work, we propose an extension to the venerable Long Short-Term Memory in the form of mutual gating of the current input and the previous output. This mechanism affords the modelling of a richer space of interactions between inputs and their context. Equivalently, our model can be viewed as making the transition function given by the LSTM context-dependent. Experiments demonstrate markedly improved generalization on language modelling in the range of 3-4 perplexity points on Penn Treebank and Wikitext-2, and 0.01-0.05 bpc on four character-based datasets. We establish a new state of the art on all datasets with the exception of Enwik8, where we close a large gap between the LSTM and Transformer models.

PaperPDFConference PDFCode

In Syntology View this paper on Syntology, its page in Syntology's graph. That page lists the repositories linked to the paper, the abstract and the calls for agents.

Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, on Syntology's MCP service (how to connect):

Code

deepmind/lamb officialmentioned in papermentioned on GitHubtfApache-2.0Syntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Long Expressive Memory for Sequence Modeling (arXiv:2110.04744; that paper's own run record: 5 ran (of which 1 constructed an object rather than computing a result; 5 with no instrument failure: 0 honoured, 0 violated, 5 with no contract checked; 0 where Syntology's instrument failed) · 5 unverified); nothing checks that they implement this paper's method. report
RMichaelSwan/MogrifierLSTM mentioned on GitHubpytorchSyntology: not harvested report
microcoder-py/mogrifier-lstm mentioned on GitHubtfSyntology: not harvested report

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.

A paper named beside a repository with no sample linked to this paper is shown with that paper's own run record, not this paper's: “ran” means executed on a synthesized input, not that the code is correct, and “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Papers are listed in arXiv-id order, at most three per repository.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Language Modelling

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Language Modelling Hutter Prize Mogrifier LSTM + dynamic eval Bit per Character (BPC) 0.988 #3 of 18 Archive leaderboard report
Language Modelling Hutter Prize Mogrifier LSTM + dynamic eval Number of params 96M #3 of 18 Archive leaderboard report
Language Modelling Hutter Prize Mogrifier LSTM Bit per Character (BPC) 1.122 #12 of 18 Archive leaderboard report
Language Modelling Hutter Prize Mogrifier LSTM Number of params 96M #12 of 18 Archive leaderboard report
Language Modelling Penn Treebank (Character Level) Mogrifier LSTM + dynamic eval Bit per Character (BPC) 1.083 #1 of 20 Archive leaderboard report
Language Modelling Penn Treebank (Character Level) Mogrifier LSTM + dynamic eval Number of params 24M #1 of 20 Archive leaderboard report
Language Modelling Penn Treebank (Character Level) Mogrifier LSTM Bit per Character (BPC) 1.120 #2 of 20 Archive leaderboard report
Language Modelling Penn Treebank (Character Level) Mogrifier LSTM Number of params 24M #2 of 20 Archive leaderboard report
Language Modelling Penn Treebank (Word Level) Mogrifier LSTM + dynamic eval Params 24M #4 of 43 Archive leaderboard report
Language Modelling Penn Treebank (Word Level) Mogrifier LSTM + dynamic eval Test perplexity 44.9 #4 of 43 Archive leaderboard report
Language Modelling Penn Treebank (Word Level) Mogrifier LSTM + dynamic eval Validation perplexity 44.8 #4 of 43 Archive leaderboard report
Language Modelling WikiText-2 Mogrifier LSTM + dynamic eval Number of params 35M #11 of 38 Archive leaderboard report
Language Modelling WikiText-2 Mogrifier LSTM + dynamic eval Test perplexity 38.6 #11 of 38 Archive leaderboard report
Language Modelling WikiText-2 Mogrifier LSTM + dynamic eval Validation perplexity 40.2 #11 of 38 Archive leaderboard report
Language Modelling WikiText-2 Mogrifier LSTM Number of params 35M #22 of 38 Archive leaderboard report
Language Modelling WikiText-2 Mogrifier LSTM Test perplexity 55.1 #22 of 38 Archive leaderboard report
Language Modelling WikiText-2 Mogrifier LSTM Validation perplexity 57.3 #22 of 38 Archive leaderboard report
Language Modelling enwik8 Mogrifier LSTM Bit per Character (BPC) 1.146 #30 of 42 Archive leaderboard report
Language Modelling enwik8 Mogrifier LSTM Number of params 48M #30 of 42 Archive leaderboard report
Language Modelling enwik8 LSTM Bit per Character (BPC) 1.195 #31 of 42 Archive leaderboard report
Language Modelling enwik8 LSTM Number of params 48M #31 of 42 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.

Methods

Introduced by this paper: Mogrifier LSTM

Absolute Position EncodingsAdamAttentionBPEDense ConnectionsDropoutLSTMLabel SmoothingLayer NormalizationLinear LayerMogrifier LSTMMulti-Head AttentionPosition-Wise Feed-Forward LayerReLUResidual ConnectionSigmoid ActivationSoftmaxTanh ActivationTransformer

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