A source content routing system is described for distributing source content received from clients such as documents, to translators for performing translation services on the source content. The routing system extracts source content features, which may be represented as vectors. The vectors may be assembled into an input matrix, which may be processed using an artificial neural network, classifier, perceptron, CRF model, and/or the like, to select a translator such as a machine translation system and/or human. The translator provides translation services translation from a source language to a target language, post translation editing, proof reading, quality analysis of a machine, quality analysis of human translation, and/or the like and returns the product to the content routing system or clients.

Methods, systems, computer program products for Program Integrated Information (PII) translation management of an application are provided. The method, according to an embodiment of the present invention, PII translation corresponding to the application of a base version is determined as PII translation of a base version by one or more processing units, and then differences between PII translation corresponding to the application of a subsequent version and PII translation of the base version is determined as PII translation of a subsequent version. Then, in a data structure, it is recorded with PII translation of the base version as a starting node of the data structure and PII translation of the subsequent version as a subsequent node of the starting node, wherein nodes in the data structure are correlated to and accessible to the application of corresponding versions.

Methods, systems, computer program products for Program Integrated Information (PII) translation management of an application are provided. The method, according to an embodiment of the present invention, PII translation corresponding to the application of a base version is determined as PII translation of a base version by one or more processing units, and then differences between PII translation corresponding to the application of a subsequent version and PII translation of the base version is determined as PII translation of a subsequent version. Then, in a data structure, it is recorded with PII translation of the base version as a starting node of the data structure and PII translation of the subsequent version as a subsequent node of the starting node, wherein nodes in the data structure are correlated to and accessible to the application of corresponding versions.

A method for translating a source text of a first language to a second language. The method includes receiving a translation request including the source text in the first language; selecting, from the source text, at least a first segment, associating at least one first metadata parameter with the first segment; providing the first segment to a first translation memory for determining a first set of translation proposals; determining a first quality score for each translation proposal; and comparing the first quality score of each translation proposal with a first predetermined acceptance threshold, and wherein based on the comparison, when a first quality score of at least one translation proposal is greater than the first predetermined acceptance threshold, the method comprises selecting a given translation proposal and providing the given translation proposal as an accepted translation of the first segment and as at least a part of an accepted translation of the source text.

A method for translating a source text of a first language to a second language. The method includes receiving a translation request including the source text in the first language; selecting, from the source text, at least a first segment, associating at least one first metadata parameter with the first segment; providing the first segment to a first translation memory for determining a first set of translation proposals; determining a first quality score for each translation proposal; and comparing the first quality score of each translation proposal with a first predetermined acceptance threshold, and wherein based on the comparison, when a first quality score of at least one translation proposal is greater than the first predetermined acceptance threshold, the method comprises selecting a given translation proposal and providing the given translation proposal as an accepted translation of the first segment and as at least a part of an accepted translation of the source text.

Disclosed are systems and methods for translation comments flowback. In some embodiments, the method includes the steps of: obtaining a first document associated with a primary document, the primary document in a primary language, the first document comprising one or more translated sections in a first language, the one or more translated sections being mapped to one or more sections in the primary document via a content identifier, the first language being different from the primary language; transmitting the first document to a first user for review; receiving a first input associated with the one or more translated sections in the first document from the first user; and populating the first input to the primary document based on the content identifier.

Computer-readable media, methods, and systems are disclosed for producing updated software documentation for a software product. A plurality of versioned source code change indications are received corresponding to changes in a plurality of source code files, which are associated with a target version of the software product. A plurality of documentation sources and associated documentation metadata is received corresponding to the plurality of versioned source code change indications. The plurality of documentation sources is merged based on the plurality of versioned source code change indications and the target version of the software product. Based on determining a set of impacted software documentation outputs, a current version of documentation build tools is fetched based on the merged plurality of merged documentation sources. Software documentation output components are built with the current version of documentation build tools. Finally, the software documentation output components are published to a primary container store.

Computer-readable media, methods, and systems are disclosed for producing updated software documentation for a software product. A plurality of versioned source code change indications are received corresponding to changes in a plurality of source code files, which are associated with a target version of the software product. A plurality of documentation sources and associated documentation metadata is received corresponding to the plurality of versioned source code change indications. The plurality of documentation sources is merged based on the plurality of versioned source code change indications and the target version of the software product. Based on determining a set of impacted software documentation outputs, a current version of documentation build tools is fetched based on the merged plurality of merged documentation sources. Software documentation output components are built with the current version of documentation build tools. Finally, the software documentation output components are published to a primary container store.

Systems and methods for automatic evaluation of the quality of NLG outputs. In some aspects of the technology, a learned evaluation model may be pretrained first using NLG model pretraining tasks, and then with further pretraining tasks using automatically generated synthetic sentence pairs. In some cases, following pretraining, the evaluation model may be further fine-tuned using a set of human-graded sentence pairs, so that it learns to approximate the grades allocated by the human evaluators.

Systems and methods for automatic evaluation of the quality of NLG outputs. In some aspects of the technology, a learned evaluation model may be pretrained first using NLG model pretraining tasks, and then with further pretraining tasks using automatically generated synthetic sentence pairs. In some cases, following pretraining, the evaluation model may be further fine-tuned using a set of human-graded sentence pairs, so that it learns to approximate the grades allocated by the human evaluators.