BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//pretalx.com//drupalcamp-kortrijk-2026//speaker//NX8QY8
BEGIN:VTIMEZONE
TZID:Europe/Brussels
BEGIN:DAYLIGHT
DTSTART:20250630T000000
TZNAME:CEST
TZOFFSETFROM:+0200
TZOFFSETTO:+0200
END:DAYLIGHT
BEGIN:STANDARD
DTSTART:20251026T030000
RDATE:20261025T030000
TZNAME:CET
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20260329T030000
RDATE:20270328T030000
TZNAME:CEST
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
SUMMARY:Reducing the Localisation Burden: Streamlining PO File Translation
  with Context-Aware LLMs - Mori Sugimoto
DTSTART;TZID=Europe/Brussels:20260630T120500
DTEND;TZID=Europe/Brussels:20260630T123000
DTSTAMP:20260811T003739Z
UID:pretalx-drupalcamp-kortrijk-2026-UFF37U@pretalx.com
DESCRIPTION:Note: This is an updated and expanded version of my previous t
 alk from DDD Athens 2026\, incorporating new features and capabilities.\n\
 nManually translating the UI text is important\, yet arduous work. While t
 he translation capability of LLMs is outstanding\, there is no guarantee t
 hat they translate new strings consistently with the existing ones.  For i
 nstance\, it may translate 'Machine name' as 'Machinale naam' instead of t
 he established translation 'Systeemnaam'.\n\nThe RAG-LLM Translator solves
  this issue by utilising existing translation strings as translation memor
 y. The architecture allows users to supply a glossary\, or generate one fr
 om existing translations.\n\nThe infrastructure is provider-agnostic\, sup
 porting both external APIs and locally deployed models. The tool also come
 s with various features such as a post-processing framework\, a RAG tuning
  helper and LLM-as-a-Judge among other builtin utilities.
LOCATION:Dropsolid room
URL:https://pretalx.com/drupalcamp-kortrijk-2026/talk/UFF37U/
END:VEVENT
END:VCALENDAR