A recoverable LLM translation system holds a 64-character validation window, resumes from boundary-safe prefixes, and passes 38 conformance tests in a sanitized artifact.
Experts, errors, and context: A large-scale study of human evalua- tion for machine translation.Transactions of the Associ- ation for Computational Linguistics, 9:1460–1474, 2021
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CL 1years
2026 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
Failure-Aware Long-Form Translation: Design and Implementation of a Recoverable LLM Translation System
A recoverable LLM translation system holds a 64-character validation window, resumes from boundary-safe prefixes, and passes 38 conformance tests in a sanitized artifact.