AI Review Agent automatically corrects errors in translations, providing a second layer of quality assurance before human review. It improves translation quality by identifying and fixing issues while enforcing your organization’s style guidelines.

​
How AI Review Works

AI Review automatically corrects translation errors and enforces style guidelines in these workflows:
  • Instant > AI Review > Review
  • Instant > AI Review > Review > Secondary Review
What AI Review checks:
  • Style guide compliance (except formatting and alignment)
  • Termbase terminology validation
  • Critical, Major, and Minor errors
What AI Review excludes:
  • Segments with “No error” status
  • Exact matches (100%) and In-context exact matches (101%)

​
Enabling AI Review

Select one of these pre-defined workflows during job creation:
  • Instant > AI Review > Review
  • Instant > AI Review > Review > Secondary Review
You can set AI Review as the default workflow at the organization or domain level. Timing: AI Review runs after Instant Translate completes and can take up to 3 hours. Users cannot translate or review segments while AI Review is running. AI Optimization: Enable this option to have error-free segments automatically confirmed and accepted at the Review stage.

Workflow and optimize with AI selectors during the job creation process.

​
Model Learning

All reviewed and accepted segments are saved to your selected model, continuing the fine-tuning of your custom models.

​
Supported Languages

AI Review is available for these language pairs:
Source languageTarget language
EnglishGerman, French, Spanish, Japanese, Korean, Portuguese, Chinese (Simplified), Chinese (Traditional), Polish, Vietnamese, Turkish, Norwegian, Finnish, Swahili, Russian, Swedish, Italian, Ukrainian, Indonesian, Dutch, Danish, Hindi, Thai, Malay, Arabic, Romanian, Czech, Hungarian
GermanEnglish
FrenchEnglish
SpanishEnglish
Language limitations:
  • Unsupported language pairs default to Instant > Review workflow
  • Jobs with languages that lack MT support cannot be submitted

​
Linguist Interface

​
Viewing AI Review Corrections

AI Review corrections appear with an “AI” segment flag. The flag color indicates error severity. Changes are shown with struck-through text (removed) in red and added text in green. You must accept segments that have been corrected by AI Review. This also reprojects any tags to ensure correct formatting.

AI Review linguist interface

​
Rejecting AI Review Corrections

You can reject AI Review changes to revert the text to its original state. When rejecting:
  1. Select a rejection category and add an optional comment
  2. The reason displays in the popover with options to edit
  3. The AI Review error icon (^) appears struck through

Rejecting an AI Review correction

​
Revision Reports

Revision reports show AI Review results alongside reviewer corrections. AI Review changes appear with “LILT AI Reviewer” comments and include the error category and severity.
Additional reviewer changes display with standard MQM error categories, severity, and comments.

​
Downloaded Revision Reports

Admin and Manager users can view the AI Review error rate percentage in downloaded Revision Reports. This calculates the error rate over the 80% of segments processed by AI Review. AI Review changes appear as:
  • Reviewer changes? column shows “Yes” for AI Review agent changes
  • Comments column shows “LILT AI Reviewer”

​
Custom Style Guide Configuration

Configure AI Review Agent with custom style guides to enforce specific requirements including:
  • Regional spelling and vocabulary (US/UK English)
  • Punctuation conventions
  • Date, time, and measurement formats
  • Language-specific formatting rules

​
Examples

​
CJK Languages

Enforce specific formatting and style rules for Chinese, Japanese, and Korean languages using authoritative sources.

​
Regional Localization

Convert between regional variations (US to UK English):
Have a linguist verify rules and results before full deployment.

​
Rule Categories

Spelling Standardization
  • Regional variations (-or/-our, -ize/-ise, -er/-re)
  • Specific word forms
Punctuation Conventions
  • Quotation mark placement
  • Abbreviation formatting
  • Serial comma usage
Date and Time Formats
  • Regional date formats (MM/DD/YYYY vs DD/MM/YYYY)
  • Time notation (12-hour vs 24-hour)
Vocabulary and Measurements
  • Regional word choices (elevator/lift)
  • Imperial vs metric conversions

​
Example Rule

Rule Name: UK_Spelling_Standardization
Command: Flag US English spellings, replace with UK equivalents.

Examples: color → colour, organize → organise, center → centre

​
Best Practices

Setup
  • Start with critical style differences for your audience
  • Use authoritative sources for style rules
  • Test with sample content before deployment
Maintenance
  • Combine AI Review with human review
  • Monitor and adjust rules based on feedback
  • Preserve technical elements ({variables}, <tags>, placeholders)

​
Advanced Grammar Rules

Collective Nouns
  • US: “The team is doing great” (singular)
  • UK: “The team are doing great” (plural)
Verb Tenses
  • US: “I just ate” (past simple)
  • UK: “I’ve just eaten” (present perfect)

​
Benefits

AI Review with custom style guides:
  • Reduces manual review time
  • Enforces consistent style guidelines
  • Scales quality assurance across projects
  • Integrates seamlessly with your workflow