Papers › Read Like Humans: Autonomous, Bidirectional and Iterative Language Modeling for Scene...

Read Like Humans: Autonomous, Bidirectional and Iterative Language Modeling for Scene Text Recognition

11 Mar 2021CVPR 2021 1arXiv:2103.06495archive 2025-07-28

Shancheng Fang, Hongtao Xie, Yuxin Wang, Zhendong Mao, Yongdong Zhang

Linguistic knowledge is of great benefit to scene text recognition. However, how to effectively model linguistic rules in end-to-end deep networks remains a research challenge. In this paper, we argue that the limited capacity of language models comes from: 1) implicitly language modeling; 2) unidirectional feature representation; and 3) language model with noise input. Correspondingly, we propose an autonomous, bidirectional and iterative ABINet for scene text recognition. Firstly, the autonomous suggests to block gradient flow between vision and language models to enforce explicitly language modeling. Secondly, a novel bidirectional cloze network (BCN) as the language model is proposed based on bidirectional feature representation. Thirdly, we propose an execution manner of iterative correction for language model which can effectively alleviate the impact of noise input. Additionally, based on the ensemble of iterative predictions, we propose a self-training method which can learn from unlabeled images effectively. Extensive experiments indicate that ABINet has superiority on low-quality images and achieves state-of-the-art results on several mainstream benchmarks. Besides, the ABINet trained with ensemble self-training shows promising improvement in realizing human-level recognition. Code is available at https://github.com/FangShancheng/ABINet.

PaperPDFConference PDFCodeCode Syntology ranMethod as code

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 28 code samples: 27 from 2 repositories linked to this paper, 1 identical to code harvested from another paper's repository. Across all 28: 18 ran (of which 11 constructed an object rather than computing a result; 13 with no instrument failure: 0 honoured, 1 violated, 12 with no contract checked; 5 where Syntology's instrument failed) · 10 unverified. Where Syntology's instrument failed: 5 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: official repository: 17 samples from 1 repository, 9 ran (7 with no instrument failure; 2 where Syntology's instrument failed); community (archive-listed): 10 samples from 1 repository, 8 ran (6 with no instrument failure; 2 where Syntology's instrument failed); 1 identical to code first harvested elsewhere. 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 on 2026-09-22 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

FangShancheng/ABINet officialmentioned in papermentioned on GitHubpytorchSyntology: 17 samples, 9 ran (2 instrument-failed) · 8 unverified report
mindspore-lab/mindocr mentioned on GitHubmindsporenot reachable when probed 2026-09-17 — repositories for recent papers often appear after camera-readySyntology: not harvested report
PaddlePaddle/PaddleOCR paddleApache-2.0Syntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to KRETA: A Benchmark for Korean Reading and Reasoning in Text-Rich VQA Attuned to... (arXiv:2508.19944; that paper's own run record: 6 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 3 violated, 0 with no contract checked; 3 where Syntology's instrument failed) · 2 unverified); nothing checks that they implement this paper's method. report
topdu/openocr pytorchApache-2.0Syntology: 10 samples, 8 ran (2 instrument-failed) · 2 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

28 samples harvested: 18 ran (of which 11 constructed an object rather than computing a result; 13 with no instrument failure: 0 honoured, 1 violated, 12 with no contract checked; 5 where Syntology's instrument failed) · 10 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 · violated contract
5ran · our draft was wrong
12ran
10unverified

Licence: 18 of the 28 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from 2 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. 1 sample is listed last, in a group without a repository. “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”. For 6 samples that ran, the graph records a signature this page does not show: for a class the fixture built without calling, the harness called the object with inputs it made up itself, so the recorded signature is the harness's probe, not the fixture's arguments. “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.

FangShancheng/ABINet official repository 17 samples, 9 ran (2 instrument-failed) · 8 unverified
BasicBlock in 126 papers
census: 6 implementations from 6 papers, 5 distinct outputs on one shared input
modules/model_abinet.py ran licence not identified · pointer only · 8b2d895aef15d9be · report fixture: returned BasicBlock · level recorded 2026-09-22
CharsetMapper modules/model_abinet.py ran licence not identified · pointer only · 94a652daa576d606 · report fixture: returned CharsetMapper · level recorded 2026-09-22
Model in 227 papers modules/model_abinet.py ran licence not identified · pointer only · 5ff3373e28aae885 · report fixture: returned Model · level recorded 2026-09-22
ResNet in 139 papers modules/model_abinet.py ran fingerprinted licence not identified · pointer only · 6233cc4fd41851f9 · report fixture: returned ResNet · level recorded 2026-09-22
TransformerDecoder in 49 papers modules/model_abinet.py ran fingerprinted licence not identified · pointer only · 226445be82e50d4d · report fixture: returned TransformerDecoder · level recorded 2026-09-22
TransformerEncoder in 96 papers modules/model_abinet.py ran fingerprinted licence not identified · pointer only · 1b76a464f3c830a7 · report fixture: returned TransformerEncoder · level recorded 2026-09-22
_get_clones in 28 papers modules/model_abinet.py ran · our draft was wrong licence not identified · pointer only · dcc52ab8e3f774c1 · report fixture: signature module:callable/0/?, N:int/0/int · returned ModuleList · level recorded 2026-09-22
multi_head_attention_forward in 73 papers modules/model_abinet.py ran · our draft was wrong licence not identified · pointer only · 62abaf6e5036ae75 · report fixture: signature query:Tensor/3/float, key:Tensor/3/float, value:Tensor/3/float, embed_dim_to_check:int/0/int, num_heads:int/0/int, in_proj_weight:Tensor/2/float, in_proj_bias:Tensor/1/float, bias_k:NoneType/0/?, bias_v:NoneType/0/?, add_zero_attn:bool/0/bool, dropout_p:float/0/float, out_proj_weight:Tensor/2/float, out_proj_bias:Tensor/1/float, training:bool/0/bool, key_padding_mask:Tensor/2/bool, need_weights:bool/0/bool, attn_mask:Tensor/2/float, use_separate_proj_weight:bool/0/bool, q_proj_weight:Tensor/2/float, k_proj_weight:Tensor/2/float, v_proj_weight:Tensor/2/float, static_k:NoneType/0/?, static_v:NoneType/0/? · returned tuple · level recorded 2026-09-22
resnet45 modules/model_abinet.py ran licence not identified · pointer only · 79aaa168270e1981 · report fixture: returned ResNet · level recorded 2026-09-22
ABINetModel modules/model_abinet.py unverified licence not identified · pointer only · 56724c2edf353bc4 · report
BCNLanguage modules/model_abinet.py unverified licence not identified · pointer only · 53798f23eb8bbe97 · report
BaseAlignment modules/model_abinet.py unverified licence not identified · pointer only · 40352875c67d1c69 · report
BaseVision modules/model_abinet.py unverified licence not identified · pointer only · b60518783b10d9a4 · report
MultiheadAttention in 44 papers modules/model_abinet.py unverified licence not identified · pointer only · 7c0b340a62978887 · report
ResTranformer modules/model_abinet.py unverified licence not identified · pointer only · afb414da038f350c · report
TransformerDecoderLayer in 45 papers modules/model_abinet.py unverified licence not identified · pointer only · cf1325762bea9012 · report
TransformerEncoderLayer in 90 papers modules/model_abinet.py unverified licence not identified · pointer only · 152fce378f96441c · report
topdu/openocr community (archive-listed) 10 samples, 8 ran (2 instrument-failed) · 2 unverified
DecoderUpsample openrec/modeling/decoders/abinet_decoder.py ran · metamorphic tier: deterministic Apache-2.0 (permissive) · b0e736ef9fcfc26b · report fixture: returned DecoderUpsample · level recorded 2026-09-22
MultiheadAttention in 44 papers openrec/modeling/decoders/abinet_decoder.py ran · metamorphic tier: deterministic Apache-2.0 (permissive) · 5dd87791ac034fb5 · report fixture: returned MultiheadAttention · level recorded 2026-09-22
PositionAttention openrec/modeling/decoders/abinet_decoder.py ran · metamorphic tier: deterministic fingerprinted Apache-2.0 (permissive) · a03b5b158fee7dbe · report fixture: returned PositionAttention · level recorded 2026-09-22
PositionalEncoding in 126 papers
census: 13 implementations from 9 papers, 9 distinct outputs across 5 buckets, one shared input per bucket
openrec/modeling/decoders/abinet_decoder.py ran · metamorphic tier: deterministic Apache-2.0 (permissive) · c02c1d1f291e2f20 · report fixture: returned PositionalEncoding · level recorded 2026-09-22
TransformerBlock in 73 papers openrec/modeling/decoders/abinet_decoder.py ran · metamorphic tier: deterministic Apache-2.0 (permissive) · 27366919489ba9ab · report fixture: returned TransformerBlock · level recorded 2026-09-22
_get_length openrec/modeling/decoders/abinet_decoder.py ran · violated contract fingerprinted Apache-2.0 (permissive) · b45dc0a4d0dae7c6 · report fixture: signature logit:Tensor/3/float · returned Tensor · level recorded 2026-09-22
_get_mask openrec/modeling/decoders/abinet_decoder.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · edacec1f0c46d5f6 · report fixture: signature length:Tensor/1/int, max_length:int/0/int · returned Tensor · level recorded 2026-09-22
encoder_layer in 6 papers openrec/modeling/decoders/abinet_decoder.py ran · our draft was wrong Apache-2.0 (permissive) · 6d0a2c5e1e8c4121 · report fixture: signature in_c:int/0/int, out_c:int/0/int, k:int/0/int, s:int/0/int, p:int/0/int · returned Sequential · level recorded 2026-09-22
ABINetDecoder openrec/modeling/decoders/abinet_decoder.py unverified Apache-2.0 (permissive) · 4fa84ab0bb6d25a1 · report
BCNLanguage openrec/modeling/decoders/abinet_decoder.py unverified Apache-2.0 (permissive) · 1e58672f91519c02 · report
Repository not located 1 sample, 1 ran (1 instrument-failed) · 0 unverified
1 sample is identical to code Syntology harvested from another paper's repository; this paper's own copy was not located, so no repository is named and no licence is recorded.
conv1x1 in 691 papers repository not located ran · our draft was wrong licence of this copy not recorded · 6ef392f365cc5703 · report fixture: signature in_planes:int/0/int, out_planes:int/0/int, stride:int/0/int · returned Conv2d · level recorded 2026-09-22

Method as code Syntology

What Syntology's graph records as this paper's claimed method, lifted into code: each assembly names its bricks (functions and classes), each with its own role and its own verification level. An assembly is not a check that the code implements the method correctly or reproduces the paper's results. Levels are each brick's own verification level in Syntology's graph (level 0: no recorded run), listed brick by brick and never combined into one level; “ran” means the brick executed on a synthesized input, not that its output is correct, and the instrument figure counts failures of Syntology's instrument, not of the code. Where a model read the code against the paper's claim, each assembly's label verdict is one model's judgement, kept in this page's JSON twin; it is not calibrated, so no card prints it. Siblings are assemblies lifted from this paper in the same batch for the same method: alternative readings kept side by side; none replaces or supersedes another.

2 assemblies on this paper: 2 harvested, 0 generated, listed in 1 groups (siblings lifted in one batch form one group).

2 sibling assemblies for ABINet (harvested), lifted in the same batch on 2026-09-09: alternative readings kept side by side; none replaces or supersedes another.

harvested Assembly for ABINet (graph method: ABINet). 11 bricks; individually on a fixture: 9 ran (of which 6 constructed an object rather than computing a result; 7 with no instrument failure: 0 honoured, 1 violated, 6 with no contract checked; 2 where Syntology's instrument failed) · 2 unverified. The assembly has not been run end-to-end.

one of 2 siblings

  1. ABINetDecoder entry · level 0 · unverified · 4fa84ab0bb6d25a1
  2. BCNLanguage model · level 0 · unverified · 1e58672f91519c02
  3. PositionAttention model · level 1 · ran · a03b5b158fee7dbe
  4. PositionalEncoding helper · level 1 · ran · c02c1d1f291e2f20
  5. TransformerBlock helper · level 1 · ran · 27366919489ba9ab
  6. _get_length helper · level 1 · ran · violated contract · b45dc0a4d0dae7c6
  7. encoder_layer helper · level 1 · ran · our draft was wrong · 6d0a2c5e1e8c4121
  8. _get_mask helper · level 1 · ran · our draft was wrong · edacec1f0c46d5f6
  9. DecoderUpsample helper · level 1 · ran · b0e736ef9fcfc26b
  10. Mlp helper · level 1 · ran · a1715055832f71fd
  11. MultiheadAttention helper · level 1 · ran · 5dd87791ac034fb5

harvested Assembly for ABINet (graph method: ABINet). 21 bricks; individually on a fixture: 13 ran (of which 7 constructed an object rather than computing a result; 8 with no instrument failure: 0 honoured, 0 violated, 8 with no contract checked; 5 where Syntology's instrument failed) · 8 unverified. The assembly has not been run end-to-end.

one of 2 siblings

  1. ABINetModel entry · level 0 · unverified · 56724c2edf353bc4
  2. BCNLanguage model · level 0 · unverified · 53798f23eb8bbe97
  3. BaseAlignment helper · level 0 · unverified · 40352875c67d1c69
  4. BaseVision model · level 0 · unverified · b60518783b10d9a4
  5. Model helper · level 1 · ran · 5ff3373e28aae885
  6. PositionalEncoding helper · level 1 · ran · 093a0d8e3c1d8ad2
  7. TransformerDecoder helper · level 1 · ran · 226445be82e50d4d
  8. TransformerDecoderLayer helper · level 0 · unverified · cf1325762bea9012
  9. ResTranformer helper · level 0 · unverified · afb414da038f350c
  10. resnet45 helper · level 1 · ran · 79aaa168270e1981
  11. CharsetMapper helper · level 1 · ran · 94a652daa576d606
  12. _get_clones helper · level 1 · ran · our draft was wrong · dcc52ab8e3f774c1
  13. MultiheadAttention helper · level 0 · unverified · 7c0b340a62978887
  14. _get_activation_fn helper · level 1 · ran · our draft was wrong · 611814a69b40fe4b
  15. TransformerEncoder helper · level 1 · ran · 1b76a464f3c830a7
  16. TransformerEncoderLayer helper · level 0 · unverified · 152fce378f96441c
  17. BasicBlock helper · level 1 · ran · 8b2d895aef15d9be
  18. ResNet helper · level 1 · ran · 6233cc4fd41851f9
  19. multi_head_attention_forward helper · level 1 · ran · our draft was wrong · 62abaf6e5036ae75
  20. conv1x1 helper · level 1 · ran · our draft was wrong · 6ef392f365cc5703
  21. conv3x3 helper · level 1 · ran · our draft was wrong · fac5364e2f53c6db

Each brick ends with its code_sha256, Syntology's identity for that exact code; click one to copy the get_code call that returns it. For agents, on Syntology's MCP service: assembly(arxiv_id="2103.06495") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

Language ModelingLanguage ModellingScene Text Recognition

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

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