Repository navigation
fix(prompt): 의미 충실 번역 — 저속/임상 표현 순화 금지 - #87
Merged
Merged
Conversation
…ical content 실통화에서 "How many times you poop?"→"몇 번이나 넣었어?", "You're so dorky"→ "도와주려고 해", "맛있어!"→"Sure" 등 오번역 관측. 원인: 프롬프트가 "ALWAYS 정중/ professional + 자연스럽게 각색 + 완곡(indirect)"를 지시해 저속·임상 표현을 순화/회피. Session A/B 템플릿에 "Faithful Meaning" 섹션 추가: - 정중함은 REGISTER(존댓말/호칭/톤)에만 적용 — 의미 변경·순화·생략·대체 금지 - 저속·임상·의료·bodily 표현도 의미 그대로 번역(검열·무관한 단어 대체 금지) - "adapt naturally"는 문장구조에만 — 의미는 정확히 유지 - 서비스/의료 통역 맥락: 의미 변조는 실제 위해 가능 회귀: prompt 관련 14 passed Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
This branch was successfully deployed
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
문제 (실통화 관측)
번역이 의미를 순화·회피:
STT는 정확(입력 모더레이션 아님) → 번역 단계에서 순화하다 의미 변조.
원인
templates.py의 지시가 **"ALWAYS 정중/professional + adapt naturally + indirect"**를 강조 → 모델이 저속·임상 표현을 "professional하게" 만들려다 의미를 뭉갬. "faithfully"는 한 줄뿐이라 압도됨.변경
Session A/B 템플릿에 "Faithful Meaning" 섹션 추가:
(예시에 노골적 표현은 넣지 않고 "clinical/bodily/blunt"로 일반화.)
검증
후속
실통화 재검증 필요 — OpenAI 자체 모더레이션이 남으면 프롬프트만으론 한계일 수 있음.