Papers › ZEN: Pre-training Chinese Text Encoder Enhanced by N-gram Representations

ZEN: Pre-training Chinese Text Encoder Enhanced by N-gram Representations

2 Nov 2019Findings of the Association for Computational Linguistics 2020arXiv:1911.00720archive 2025-07-28

Shizhe Diao, Jiaxin Bai, Yan Song, Tong Zhang, Yonggang Wang

The pre-training of text encoders normally processes text as a sequence of tokens corresponding to small text units, such as word pieces in English and characters in Chinese. It omits information carried by larger text granularity, and thus the encoders cannot easily adapt to certain combinations of characters. This leads to a loss of important semantic information, which is especially problematic for Chinese because the language does not have explicit word boundaries. In this paper, we propose ZEN, a BERT-based Chinese (Z) text encoder Enhanced by N-gram representations, where different combinations of characters are considered during training. As a result, potential word or phase boundaries are explicitly pre-trained and fine-tuned with the character encoder (BERT). Therefore ZEN incorporates the comprehensive information of both the character sequence and words or phrases it contains. Experimental results illustrated the effectiveness of ZEN on a series of Chinese NLP tasks. We show that ZEN, using less resource than other published encoders, can achieve state-of-the-art performance on most tasks. Moreover, it is shown that reasonable performance can be obtained when ZEN is trained on a small corpus, which is important for applying pre-training techniques to scenarios with limited data. The code and pre-trained models of ZEN are available at https://github.com/sinovation/zen.

PaperPDFConference PDFCodeCode Syntology ran

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also 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

Syntology 35 code samples harvested from 5 repositories linked to this paper: 28 ran (of which 0 constructed an object rather than computing a result; 22 with no instrument failure: 1 honoured, 0 violated, 21 with no contract checked; 6 where Syntology's instrument failed) · 7 unverified. Where Syntology's instrument failed: 6 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: official repository: 2 samples from 1 repository, 1 ran (0 with no instrument failure; 1 where Syntology's instrument failed); community (archive-listed): 33 samples from 4 repositories, 27 ran (22 with no instrument failure; 5 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

No sample is linked to this paper from 1 of its repositories that Syntology harvested for other papers:

Levels written between 2026-09-22 and 2026-09-25 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed. 2 of the 28 samples that ran have no recorded date.

sinovation/ZEN officialmentioned in papermentioned on GitHubpytorchSyntology: 2 samples, 1 ran (1 instrument-failed) · 1 unverified report
SVAIGBA/TwASP mentioned on GitHubpytorchMITSyntology: 17 samples, 13 ran (5 instrument-failed) · 4 unverified report
SVAIGBA/WMSeg mentioned on GitHubpytorchMITSyntology: 8 samples, 7 ran (0 instrument-failed) · 1 unverified report
YYGe01/ZEN mentioned on GitHubpytorchnot reachable when probed 2026-09-17 — repositories for recent papers often appear after camera-readySyntology: not harvested report
cuhksz-nlp/SAPar mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Improving Constituency Parsing with Span Attention (arXiv:2010.07543; that paper's own run record: 1 ran (of which 1 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified; the one sample that ran constructed an object rather than computing a result); nothing checks that they implement this paper's method. report
cuhksz-nlp/het-mc mentioned on GitHubpytorchMITSyntology: 4 samples, 3 ran (0 instrument-failed) · 1 unverified report
cuhksz-nlp/mcasp mentioned on GitHubpytorchMITSyntology: 4 samples, 4 ran (0 instrument-failed) · 0 unverified 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

35 samples harvested: 28 ran (of which 0 constructed an object rather than computing a result; 22 with no instrument failure: 1 honoured, 0 violated, 21 with no contract checked; 6 where Syntology's instrument failed) · 7 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page has not changed since , the first build that kept a record date for it; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

1ran · honoured contract
6ran · our draft was wrong
21ran
7unverified

Licence: 2 of the 35 samples are pointer only, meaning Syntology does not serve that copy's text. For 2 of them, this paper's copy is under a permissive licence and cleared, but the stored copy of the same code that Syntology serves is not; the licence label's tooltip gives Syntology's recorded reason. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from 5 repositories linked to this paper, official or community; the list below is grouped by repository, each group headed by its name and what it is to the paper. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of six reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity. A census tag appears when Syntology's census ran implementations with that name on one shared input per bucket: it gives how many were compared, from how many papers, and how many distinct outputs they produced, and names this sample's bucket when this sample was one of them.

sinovation/ZEN official repository 2 samples, 1 ran (1 instrument-failed) · 1 unverified
cws_evaluate_word_PRF examples/run_token_level_classification.py ran · our draft was wrong Apache-2.0 (permissive) · ba2b25faa32a1249 · report fixture: signature y_pred:str/0/str, y:str/0/str · returned dict · level recorded 2026-09-22
convert_example_to_features in 10 papers examples/run_pre_train.py unverified Apache-2.0 (permissive) · 000595907a687ec4 · report
SVAIGBA/TwASP community (archive-listed) 17 samples, 13 ran (5 instrument-failed) · 4 unverified
swish in 258 papers
census: 14 implementations from 37 papers, 2 distinct outputs on one shared input this sample: bucket 1 of 1, arg 1: rank 2, kind float, dtype float32
pytorch_pretrained_bert/modeling_openai.py ran · our draft was wrong fingerprinted MIT (permissive) · 0f786c407fb1ee4c · report fixture: signature x:Tensor/2/float · returned Tensor · level recorded, date not recorded
build_tf_to_pytorch_map in 14 papers pytorch_pretrained_bert/modeling_transfo_xl.py ran · our draft was wrong MIT (permissive) · 35ef111fc00c21b4 · report fixture: signature model:callable/0/?, config:Config/0/? · returned dict · level recorded 2026-09-22
eval_sentence twasp_eval.py ran MIT (permissive) · 87867d95beeee94a · report fixture: signature y_pred:list/1/str, y:list/1/str, sentence:str/0/str, word2id:dict/0/? · returned tuple · level recorded 2026-09-25
gelu in 308 papers
census: 43 implementations from 74 papers, 6 distinct outputs on one shared input this sample: bucket 1 of 1, arg 1: rank 2, kind float, dtype float32
pytorch_pretrained_bert/modeling_gpt2.py ran · our draft was wrong fingerprinted MIT (permissive) · 8d23fbe2b99b840b · report fixture: signature x:Tensor/2/float · returned Tensor · level recorded, date not recorded
gelu in 308 papers
census: 43 implementations from 74 papers, 6 distinct outputs on one shared input this sample: bucket 1 of 1, arg 1: rank 2, kind float, dtype float32
pytorch_pretrained_bert/modeling.py ran · honoured contract fingerprinted MIT (permissive) here; the stored copy is pointer only · fdc64f4c72036ae4 · report fixture: signature x:Tensor/2/float · returned Tensor · level recorded 2026-09-22
get_word2id twasp_helper.py ran MIT (permissive) · 98046cec1e6090e5 · report fixture: signature train_path:str/0/str · returned dict · level recorded 2026-09-25
pos_evaluate_OOV twasp_eval.py ran MIT (permissive) · 342218755bdf47ed · report fixture: signature y_pred_list:list/1/list, y_list:list/1/list, sentence_list:list/1/list, word2id:dict/0/? · returned tuple · level recorded 2026-09-25
prune_conv1d_layer in 20 papers pytorch_pretrained_bert/modeling_gpt2.py ran MIT (permissive) · f255dd42f46c11c1 · report fixture: signature layer:callable/0/?, index:Tensor/1/int, dim:int/0/int · returned Conv1D · level recorded 2026-09-25
prune_linear_layer in 40 papers pytorch_pretrained_bert/modeling.py ran · our draft was wrong MIT (permissive) · d30d6c3098df2c42 · report fixture: signature layer:callable/0/?, index:Tensor/1/int, dim:int/0/int · returned Linear · level recorded 2026-09-22
read_sentence twasp_model.py ran MIT (permissive) · 5b7c711347a2c214 · report fixture: signature filename:str/0/str · returned list · level recorded 2026-09-25
read_tsv in 13 papers twasp_helper.py ran MIT (permissive) · 352e22ef30c1f7b7 · report fixture: signature file_path:str/0/str · returned tuple · level recorded 2026-09-25
readfile in 11 papers twasp_model.py ran MIT (permissive) · 013a1a80d4e2be45 · report fixture: signature filename:str/0/str · returned list · level recorded 2026-09-25
sample_logits in 16 papers pytorch_pretrained_bert/modeling_transfo_xl_utilities.py ran · our draft was wrong MIT (permissive) · 93228a3eb5c4d179 · report fixture: signature embedding:callable/0/?, bias:Tensor/1/float, labels:Tensor/2/int, inputs:Tensor/3/float, sampler:LogUniformSampler/0/? · returned Tensor · level recorded 2026-09-22
load_tf_weights_in_bert in 135 papers pytorch_pretrained_bert/modeling.py unverified MIT (permissive) here; the stored copy is pointer only · b1a7180868dd8d6a · report
load_tf_weights_in_gpt2 in 68 papers pytorch_pretrained_bert/modeling_gpt2.py unverified MIT (permissive) · a62604f618a8cc82 · report
load_tf_weights_in_openai_gpt in 13 papers pytorch_pretrained_bert/modeling_openai.py unverified MIT (permissive) · 46f305675277e757 · report
read_sentence twasp_helper.py unverified MIT (permissive) · 4d322567015518e5 · report
SVAIGBA/WMSeg community (archive-listed) 8 samples, 7 ran (0 instrument-failed) · 1 unverified
count_n_gram wmseg_helper.py ran MIT (permissive) · 4892acce036705d1 · report fixture: signature data_dir:str/0/str · returned dict · level recorded 2026-09-25
cws_evaluate_OOV wmseg_eval.py ran MIT (permissive) · 1a310e8f6b36323b · report fixture: signature y_pred_list:list/1/list, y_list:list/1/list, sentence_list:list/1/list, word2id:dict/0/? · returned float · level recorded 2026-09-25
cws_evaluate_word_PRF wmseg_eval.py ran MIT (permissive) · 0421eab98cc3b9a9 · report fixture: signature y_pred:str/0/str, y:str/0/str · returned tuple · level recorded 2026-09-25
eval_sentence wmseg_eval.py ran MIT (permissive) · fad09e6582176e41 · report fixture: signature y_pred:list/1/str, y:list/1/str, sentence:str/0/str, word2id:dict/0/? · returned tuple · level recorded 2026-09-25
read_tsv in 13 papers wmseg_helper.py ran MIT (permissive) · efebd15616e8be5b · report fixture: signature file_path:str/0/str · returned tuple · level recorded 2026-09-25
readfile in 11 papers wmseg_model.py ran MIT (permissive) · 3e642a8045e6ba92 · report fixture: signature filename:str/0/str, flag:str/0/str · returned list · level recorded 2026-09-25
readsentence wmseg_model.py ran MIT (permissive) · 9b29b0de8fad09f9 · report fixture: signature filename:str/0/str · returned list · level recorded 2026-09-25
dlg wmseg_helper.py unverified MIT (permissive) · 301c6b5f8def3f58 · report
cuhksz-nlp/het-mc community (archive-listed) 4 samples, 3 ran (0 instrument-failed) · 1 unverified
get_vocab in 15 papers hetmc_helper.py ran MIT (permissive) · 99b87b401d1fae1e · report fixture: signature training_dialog:list/1/tuple · returned tuple · level recorded 2026-09-24
load_json in 344 papers hetmc_helper.py ran MIT (permissive) · 3872690ce9ded1c4 · report fixture: signature file_path:str/0/str · returned dict · level recorded 2026-09-24
read_dialog hetmc_helper.py ran MIT (permissive) · c9622c7a1d0912d6 · report fixture: signature filename:str/0/str · returned list · level recorded 2026-09-24
load_tf_weights_in_transfo_xl in 12 papers pytorch_pretrained_bert/modeling_transfo_xl.py unverified MIT (permissive) · 1ebc2bf2c3cddc4d · report
cuhksz-nlp/mcasp community (archive-listed) 4 samples, 4 ran (0 instrument-failed) · 0 unverified
dlg mcasp_helper.py ran MIT (permissive) · 34ec071d80ec7278 · report fixture: signature data_path_list:list/1/str, ngram_length:int/0/int, renew_freq:int/0/int · returned dict · level recorded 2026-09-24
eval_sentence mcasp_eval.py ran MIT (permissive) · 116ccd8fe121ef8f · report fixture: signature y_pred:list/1/str, y:list/1/str, sentence:str/0/str, word2id:dict/0/? · returned tuple · level recorded 2026-09-24
get_word2id mcasp_helper.py ran MIT (permissive) · ecd45f0d33c0aff5 · report fixture: signature train_path:str/0/str · returned dict · level recorded 2026-09-24
pos_evaluate_OOV mcasp_eval.py ran MIT (permissive) · 3f4a9392fcb8695c · report fixture: signature y_pred_list:list/1/list, y_list:list/1/list, sentence_list:list/1/list, word2id:dict/0/? · returned tuple · level recorded 2026-09-24

Tasks

Chinese Named Entity RecognitionChinese Word SegmentationDocument ClassificationNatural Language InferencePart-Of-Speech TaggingSentence Pair ModelingSentiment Analysis

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Chinese Named Entity Recognition MSRA ZEN (Init with Chinese BERT) F1 95.25 #9 of 21 Archive leaderboard report
Chinese Named Entity Recognition MSRA ZEN (Random Init) F1 93.24 #18 of 21 Archive leaderboard report
Chinese Word Segmentation MSR ZEN (Init with Chinese BERT) F1 98.35 #4 of 6 Archive leaderboard report
Chinese Word Segmentation MSR ZEN (Random Init) F1 97.89 #6 of 6 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