Papers › CANINE: Pre-training an Efficient Tokenization-Free Encoder for Language Representation

CANINE: Pre-training an Efficient Tokenization-Free Encoder for Language Representation

11 Mar 2021arXiv:2103.06874archive 2025-07-28

Jonathan H. Clark, Dan Garrette, Iulia Turc, John Wieting

Pipelined NLP systems have largely been superseded by end-to-end neural modeling, yet nearly all commonly-used models still require an explicit tokenization step. While recent tokenization approaches based on data-derived subword lexicons are less brittle than manually engineered tokenizers, these techniques are not equally suited to all languages, and the use of any fixed vocabulary may limit a model's ability to adapt. In this paper, we present CANINE, a neural encoder that operates directly on character sequences, without explicit tokenization or vocabulary, and a pre-training strategy that operates either directly on characters or optionally uses subwords as a soft inductive bias. To use its finer-grained input effectively and efficiently, CANINE combines downsampling, which reduces the input sequence length, with a deep transformer stack, which encodes context. CANINE outperforms a comparable mBERT model by 2.8 F1 on TyDi QA, a challenging multilingual benchmark, despite having 28% fewer model parameters.

PaperPDFCode

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

Code

huggingface/transformers mentioned on GitHubpytorch report
octanove/shiba mentioned on GitHubpytorchNOASSERTION report
pwc-1/Paper-8 mindspore 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.

Code Syntology ran Syntology

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

Tasks

Inductive Bias

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

Absolute Position EncodingsAdamAttentionBPECANINEConvolutionDense ConnectionsDropoutLabel SmoothingLayer NormalizationLinear LayerMulti-Head AttentionPosition-Wise Feed-Forward LayerResidual ConnectionSoftmaxTransformermBERT

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