context-aware correction

as donto believed on

From Dontopedia, the open, paraconsistent wiki. (Last updated 2026-06-11.)

context-aware correction has 85 facts recorded in Dontopedia across 10 references, with 14 live disagreements.

85 facts·35 predicates·10 sources·14 in dispute

Mostly:rdf:type(10), has sequential steps(9), performs sequence(7)

Rdf:typein dispute`rdf:type`

Correction Approach^[1]all time · 14ffc028 Ee6d 42c4 B485 Bab0210f90c7
Technique^[3]all time · F05bdfec F74c 4a81 91da F88d561731be
Search Technique^[3]all time · F05bdfec F74c 4a81 91da F88d561731be
Correction Technique^[4]sourceall time · F9c8a1fd 99fa 42bd Aafa D15a41dbfd3c
Function^[5]all time · A8d4e00d 0adb 49c2 A304 E8356b9d69a3
Component^[6]all time · 5463aea7 1918 406e 92aa D3bd2fc59518
Function^[7]all time · 3b8e94e6 6ea2 40ce B7fd Ddc4e92b2865
Function^[8]all time · 4346daa8 69e0 41ac A434 F64d60c67428
Capability^[9]all time · Fee22513 6932 45df 8fbd 48ecb3f71f7f
Correction Method^[10]all time · Cd1202e2 8ff4 46e7 B33d 4ac9df22522f

Inbound mentions (27)

Other subjects in dontopedia point AT this entity as a value. These are inverse relationships — e.g. "X motherOf this subject" — and answer questions the forward facts can't. Grouped by predicate.

`combines`Combines(2)

Combined Approachex:combined-approach
Hybrid Approachex:hybrid-approach

`contains`Contains(2)

Hybrid Sectionex:hybrid-section
Implementation Stepsex:implementation-steps

`hasComponent`Has Component(2)

Combined Approachex:combined-approach
Spelling Correction Systemex:spelling-correction-system

`isUsedBy`Is Used by(2)

Bert Modelex:bert-model
Tokenizerex:tokenizer

`usedBy`Used by(2)

Tokenizerex:tokenizer
Torchex:torch

`calls`Calls(1)

Spell Correction With Cacheex:spell-correction-with-cache

`capableOf`Capable of(1)

Machine Learning Modelex:machine-learning-model

`containsFunction`Contains Function(1)

Scriptex:script

`describes`Describes(1)

Source Documentex:source-document

`enables`Enables(1)

Bert Modelex:bert-model

`hasImplementationStep`Has Implementation Step(1)

Spelling Correction Systemex:spelling-correction-system

`hasOption`Has Option(1)

Fallback Mechanismex:fallback-mechanism

`hasPart`Has Part(1)

Scriptex:script

`hasTask`Has Task(1)

Week 1ex:week-1

`implements`Implements(1)

Spelling Correction Classex:spelling-correction-class

`isParameterOf`Is Parameter of(1)

Input Text Parameterex:input-text-parameter

`providesInstructionsFor`Provides Instructions for(1)

Source Documentex:source-document

`purposeOf`Purpose of(1)

Correct With Contextex:correct-with-context

`requiredBy`Required by(1)

Use Transformers Libraryex:use-transformers-library

`teaches`Teaches(1)

Source Documentex:source-document

`usedFor`Used for(1)

Bert Modelex:bert-model

`uses`Uses(1)

Spell Correction Logicex:spell-correction-logic

Other facts (71)

The long tail: predicates that appear too rarely to warrant their own section. Filter or scroll to find a specific one. Each row links to its source.

71 facts

Predicate	Value	Ref
Has Sequential Steps	Tokenization Step	[5]
Has Sequential Steps	Masking Step	[5]
Has Sequential Steps	Id Conversion Step 1	[5]
Has Sequential Steps	Tensor Conversion Step	[5]
Has Sequential Steps	Model Inference Step	[5]
Has Sequential Steps	Prediction Extraction Step	[5]
Has Sequential Steps	Argmax Step	[5]
Has Sequential Steps	Token Conversion Step	[5]
Has Sequential Steps	Id Conversion Step 2	[5]
Performs Sequence	Tokenization Sequence	[5]
Performs Sequence	Masking Sequence	[5]
Performs Sequence	Id Conversion Sequence	[5]
Performs Sequence	Tensor Conversion Sequence	[5]
Performs Sequence	Inference Sequence	[5]
Performs Sequence	Prediction Sequence	[5]
Performs Sequence	Argmax Sequence	[5]
Calls	tokenizer.tokenize	[5]
Calls	tokenizer.convert_tokens_to_ids	[5]
Calls	torch.tensor	[5]
Calls	model(masked_input)	[5]
Calls	torch.argmax	[5]
Calls	tokenizer.convert_ids_to_tokens	[5]
Assigns	tokens	[5]
Assigns	masked_input	[5]
Assigns	outputs	[5]
Assigns	predicted_indices	[5]
Assigns	corrected_tokens	[5]
Assigns	corrected_text	[5]
Indexes	Outputs	[5]
Indexes	Predicted Indices	[5]
Indexes	Outputs Variable	[5]
Indexes	Predicted Indices Variable	[5]
Uses	pre-trained-language-model	[2]
Uses	Tokenizer	[5]
Uses	Pre Trained Bert Model	[6]
Imports	Bert Tokenizer	[7]
Imports	Bert for Masked Lm	[7]
Imports	Torch	[7]
Has Parameter	input_text	[5]
Has Parameter	input_text	[8]
Wraps in	torch.no_grad	[5]
Wraps in	No Grad Context	[5]
Returns	Corrected Text	[5]
Returns	corrected_text	[8]
Calls Method	Convert Tokens to Ids Method	[5]
Calls Method	Convert Ids to Tokens Method	[5]
Purpose	Perform Context Aware Correction	[6]
Purpose	correct-with-context	[8]
Consists of	Masking Step	[6]
Consists of	Prediction Step	[6]
Described in	Source Document	[1]
Requires	Pre Trained Language Model	[1]
Implemented by	Spelling Correction Class	[1]
Timeframe	Day 5-7	[2]
Action	set-up-context-aware-correction	[2]
Enabled by	Bert Model	[4]
Is Enabled by	Bert Model	[4]
Is Part of	Combined Approach	[4]
Has Name	context_aware_correction	[5]
Has Docstring	Perform context-aware correction using BERT.	[5]
Uses List Comprehension	token if token.isalpha() else '[MASK]'	[5]
Defined in	Script	[5]
Uses Bert Model	Bert Model	[5]
Uses Tokenizer	Tokenizer	[5]
Creates Tensor	Masked Input	[5]
Applies Arg Max	Predictions	[5]
Has Assignment	Corrected Text Variable	[5]
Enables	Predict Correct Tokens	[6]
Complements	Dictionary Lookups	[6]
Parameter	Input Text	[7]
Called by	Spell Correction With Cache	[8]

Timeline axis is valid_time — when each source says the fact was true in the world, not when Dontopedia learned about it. Retracted rows are kept for provenance; coloured stripes indicate the context kind.

85 undated

—

typebeam/14ffc028-ee6d-42c4-b485-bab0210f90c7

ex:CorrectionApproach

—

describedInbeam/14ffc028-ee6d-42c4-b485-bab0210f90c7

ex:source-document

—

requiresbeam/14ffc028-ee6d-42c4-b485-bab0210f90c7

ex:pre-trained-language-model

—

implementedBybeam/14ffc028-ee6d-42c4-b485-bab0210f90c7

ex:spelling-correction-class

—

timeframebeam/c249ccfb-cea0-44d2-b952-eb744cad24ed

Day 5-7

—

actionbeam/c249ccfb-cea0-44d2-b952-eb744cad24ed

set-up-context-aware-correction

—

usesbeam/c249ccfb-cea0-44d2-b952-eb744cad24ed

pre-trained-language-model

—

typebeam/f05bdfec-f74c-4a81-91da-f88d561731be

ex:Technique

—

labelbeam/f05bdfec-f74c-4a81-91da-f88d561731be

Context-aware corrections

—

typebeam/f05bdfec-f74c-4a81-91da-f88d561731be

ex:SearchTechnique

—

typebeam/f9c8a1fd-99fa-42bd-aafa-d15a41dbfd3c

ex:CorrectionTechnique

—

labelbeam/f9c8a1fd-99fa-42bd-aafa-d15a41dbfd3c

context-aware correction

—

enabledBybeam/f9c8a1fd-99fa-42bd-aafa-d15a41dbfd3c

ex:bert-model

—

isEnabledBybeam/f9c8a1fd-99fa-42bd-aafa-d15a41dbfd3c

ex:bert-model

—

isPartOfbeam/f9c8a1fd-99fa-42bd-aafa-d15a41dbfd3c

ex:combined-approach

—

typebeam/a8d4e00d-0adb-49c2-a304-e8356b9d69a3

ex:Function

—

hasNamebeam/a8d4e00d-0adb-49c2-a304-e8356b9d69a3

context_aware_correction

—

hasParameterbeam/a8d4e00d-0adb-49c2-a304-e8356b9d69a3

input_text

—

hasDocstringbeam/a8d4e00d-0adb-49c2-a304-e8356b9d69a3

Perform context-aware correction using BERT.

—

callsbeam/a8d4e00d-0adb-49c2-a304-e8356b9d69a3

tokenizer.tokenize

—

assignsbeam/a8d4e00d-0adb-49c2-a304-e8356b9d69a3

tokens

—

assignsbeam/a8d4e00d-0adb-49c2-a304-e8356b9d69a3

masked_input

—

usesListComprehensionbeam/a8d4e00d-0adb-49c2-a304-e8356b9d69a3

token if token.isalpha() else '[MASK]'

—

callsbeam/a8d4e00d-0adb-49c2-a304-e8356b9d69a3

tokenizer.convert_tokens_to_ids

—

callsbeam/a8d4e00d-0adb-49c2-a304-e8356b9d69a3

torch.tensor

—

wrapsInbeam/a8d4e00d-0adb-49c2-a304-e8356b9d69a3

torch.no_grad

—

callsbeam/a8d4e00d-0adb-49c2-a304-e8356b9d69a3

model(masked_input)

—

assignsbeam/a8d4e00d-0adb-49c2-a304-e8356b9d69a3

outputs

—

callsbeam/a8d4e00d-0adb-49c2-a304-e8356b9d69a3

torch.argmax

—

assignsbeam/a8d4e00d-0adb-49c2-a304-e8356b9d69a3

predicted_indices

—

callsbeam/a8d4e00d-0adb-49c2-a304-e8356b9d69a3

tokenizer.convert_ids_to_tokens

—

assignsbeam/a8d4e00d-0adb-49c2-a304-e8356b9d69a3

corrected_tokens

—

assignsbeam/a8d4e00d-0adb-49c2-a304-e8356b9d69a3

corrected_text

—

definedInbeam/a8d4e00d-0adb-49c2-a304-e8356b9d69a3

ex:script

—

usesBERTModelbeam/a8d4e00d-0adb-49c2-a304-e8356b9d69a3

ex:bert-model

—

performsSequencebeam/a8d4e00d-0adb-49c2-a304-e8356b9d69a3

ex:tokenization-sequence

—

performsSequencebeam/a8d4e00d-0adb-49c2-a304-e8356b9d69a3

ex:masking-sequence

—

performsSequencebeam/a8d4e00d-0adb-49c2-a304-e8356b9d69a3

ex:id-conversion-sequence

—

performsSequencebeam/a8d4e00d-0adb-49c2-a304-e8356b9d69a3

ex:tensor-conversion-sequence

—

performsSequencebeam/a8d4e00d-0adb-49c2-a304-e8356b9d69a3

ex:inference-sequence

—

performsSequencebeam/a8d4e00d-0adb-49c2-a304-e8356b9d69a3

ex:prediction-sequence

—

usesbeam/a8d4e00d-0adb-49c2-a304-e8356b9d69a3

ex:tokenizer

—

returnsbeam/a8d4e00d-0adb-49c2-a304-e8356b9d69a3

ex:corrected-text

—

wrapsInbeam/a8d4e00d-0adb-49c2-a304-e8356b9d69a3

ex:no-grad-context

—

performsSequencebeam/a8d4e00d-0adb-49c2-a304-e8356b9d69a3

ex:argmax-sequence

—

usesTokenizerbeam/a8d4e00d-0adb-49c2-a304-e8356b9d69a3

ex:tokenizer

—

callsMethodbeam/a8d4e00d-0adb-49c2-a304-e8356b9d69a3

ex:convert-tokens-to-ids-method

—

callsMethodbeam/a8d4e00d-0adb-49c2-a304-e8356b9d69a3

ex:convert-ids-to-tokens-method

—

createsTensorbeam/a8d4e00d-0adb-49c2-a304-e8356b9d69a3

ex:masked-input

—

appliesArgMaxbeam/a8d4e00d-0adb-49c2-a304-e8356b9d69a3

ex:predictions

—

indexesbeam/a8d4e00d-0adb-49c2-a304-e8356b9d69a3

ex:outputs

—

indexesbeam/a8d4e00d-0adb-49c2-a304-e8356b9d69a3

ex:predicted-indices

—

indexesbeam/a8d4e00d-0adb-49c2-a304-e8356b9d69a3

ex:outputs-variable

—

indexesbeam/a8d4e00d-0adb-49c2-a304-e8356b9d69a3

ex:predicted-indices-variable

—

hasAssignmentbeam/a8d4e00d-0adb-49c2-a304-e8356b9d69a3

ex:corrected-text-variable

—