Governing Two Primary LLM Review Streams and a Separate GPT-5 Audit in Evidence Synthesis: Source-Based Human Reconciliation in a Bangla Visual Speech Recognition Case Study
Publications · 02.11.2026 · 14 (4) · 74
DOI: 10.3390/publications14040074 ↗Kurz gefasst
Fachartikel in der Zeitschrift Publications, online erschienen am 02.11.2026. Verfasst von A. Hasib Uddin und 1 weiteren Autor:innen. Die Arbeit nennt 44 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Division of Information and Communication Technology (ICT), Ministry of Posts, Telecommunications and Information Technology, Government of the People’s Republ….
Abstract (Originalsprache)
Large language models (LLMs) are increasingly used in evidence synthesis, but model agreement cannot substitute for source verification. We retrospectively examine a human-governed workflow embedded in a Bangla visual speech recognition evidence map. Two primary LLM streams assessed the same paired records in separate sessions using retained workbook instructions. The paired pool contained 17 records: 16 Bangla/Bengali records and one Assamese record (BA-017). After restriction to Bangla/Bengali by language scope, 16 paired records remained; all nine included Bangla reports were retrieved in full. Exact title/abstract decisions matched for 16/17 records in the full pool (94.1%; kappa = 0.887) and 15/16 in the Bangla set (93.8%; kappa = 0.881). Operational retrieve decisions matched for all records in both sets, and both streams identified the same nine included and seven excluded Bangla records. Separately, GPT-5 recoded 81 post hoc source-verifiable audit cells; 76 matched the reconciled provisional ledger (93.8%). Humans source-checked all 81 cells, and the five discordant provisional values changed after source adjudication. Because adjudication was not blinded to discrepancy status, the audit comparison is not an accuracy or independent-validation estimate. This bounded case contributes a role-and-record framework for auditable LLM-assisted evidence synthesis while retaining human authority over final source-based decisions.
Förderung
- Division of Information and Communication Technology (ICT), Ministry of Posts, Telecommunications and Information Technology, Government of the People’s Republ…
Zitieren
APA
Uddin, A. H., & Arif, A. S. M. (2026). Governing Two Primary LLM Review Streams and a Separate GPT-5 Audit in Evidence Synthesis: Source-Based Human Reconciliation in a Bangla Visual Speech Recognition Case Study. Publications, 14(4), 74. https://doi.org/10.3390/publications14040074
Vancouver
Uddin AH, Arif ASM. Governing Two Primary LLM Review Streams and a Separate GPT-5 Audit in Evidence Synthesis: Source-Based Human Reconciliation in a Bangla Visual Speech Recognition Case Study. Publications. 2026;14(4):74. doi:10.3390/publications14040074
BibTeX
@article{Uddin2026,
author = {Uddin, A. Hasib and Arif, Abu Shamim Mohammad},
title = {Governing Two Primary LLM Review Streams and a Separate GPT-5 Audit in Evidence Synthesis: Source-Based Human Reconciliation in a Bangla Visual Speech Recognition Case Study},
journal = {Publications},
year = {2026},
volume = {14},
number = {4},
pages = {74},
doi = {10.3390/publications14040074}
}
EUVIMED