Everything here is rule-based: values are substituted into your own approved file, and every finding is proved from the text itself. The same input always gives the same result.
What it has learned here, and teaching it a new type from approved examples.
Copy of Experience CertificateTamurakoma & Co., Ltd.
7 fields · learned from 3 examples
Copy of ExperienceNippon Koei Bangladesh
4 fields · learned from 3 examples
Declaration for foreign employeeTamurakoma & Co., Ltd.
6 fields · learned from 6 examples
Recommendation LetterTamurakoma & Co., Ltd.
15 fields · learned from 8 examples
Invitation LetterTamurakoma & Co., Ltd.
11 fields · learned from 9 examples
Authorization LetterMORIRIN CO., LTD.
11 fields · learned from 2 examples
Forwarding Letter-RecommendationMORIRIN CO., LTD.
11 fields · learned from 3 examples
Invitation Letter to Philippine Immigration (Japan Inspection Co., Ltd.
7 fields · learned from 2 examples
Invitation Letter to Applicant (Japan Inspection Co., Ltd.
4 fields · learned from 2 examples
CV(Bio Data) of Watanabe SanJapan Inspection Co., Ltd.
2 fields · learned from 2 examples
Certification from TCF BangladeshJapan Inspection Co., Ltd.
3 fields · learned from 2 examples
Boi dataTS Tech
15 fields · learned from 3 examples
BIO DataTamurakoma & Co., Ltd.
1 field · learned from 2 examples
Experience CertificateTamurakoma & Co., Ltd.
7 fields · learned from 2 examples
Appointment letterNippon Koei Bangladesh
6 fields · learned from 2 examples
Board resolutionNippon Koei Bangladesh
2 fields · learned from 2 examples
Pick two or more approved documents of the same kind that differ only in their details — three notices for different shareholders, say. Comparing them shows which parts are the format and which are the facts. The first one you tick is the base, and every draft is built from that exact file, so drafts keep its letterhead and layout.