koriweb
95cd8bb891
- 코드 그라운딩: 기술 주제 문서의 '적용 사례'에 실제 레포 구현 위치
(file:line)+커밋 자동 주입 (예: 문서 청킹 전략→connectai/src/retrieval/chunker.ts).
멱등 마커(CODE-GROUNDING)로 재실행 시 갱신.
- MOC: 39개 클러스터 폴더에 _MOC.md 학습지도 생성(진입점+통찰 주석).
도구: Datacollect/scripts/{code_grounding,moc_generator}.mjs
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
163 lines
5.2 KiB
Markdown
163 lines
5.2 KiB
Markdown
---
|
|
id: wiki-2026-0508-self-verification
|
|
title: Self-Verification
|
|
category: 10_Wiki/Topics
|
|
status: verified
|
|
canonical_id: self
|
|
aliases: [Chain of Verification, CoVe, Self-Critique, LLM Self-Check]
|
|
duplicate_of: none
|
|
source_trust_level: A
|
|
confidence_score: 0.9
|
|
verification_status: applied
|
|
tags: [llm, reasoning, prompting, reliability]
|
|
raw_sources: []
|
|
last_reinforced: 2026-05-10
|
|
github_commit: pending
|
|
tech_stack:
|
|
language: python
|
|
framework: Anthropic SDK / OpenAI SDK
|
|
---
|
|
|
|
# Self-Verification
|
|
|
|
## 매 한 줄
|
|
> **"매 LLM 이 자기 답을 다시 점검 — generate → verify → revise"**. 매 Dhuliawala 2023 의 CoVe (Chain of Verification), self-consistency, self-refine, reflexion 가 매 family. 매 2026: reasoning model (Claude Opus 4.7 thinking, o3) 이 매 internalized self-verify, 그래도 매 explicit verify pass 가 critical accuracy 추가.
|
|
|
|
## 매 핵심
|
|
|
|
### 매 형태
|
|
- **Self-consistency** (Wang 2022): 매 sample N 개 → majority vote.
|
|
- **Chain of Verification** (CoVe): plan → baseline → verify Qs → answer Qs → final.
|
|
- **Self-refine** (Madaan 2023): generate → critique → revise loop.
|
|
- **Reflexion**: episodic memory of past mistakes.
|
|
- **Constitutional / RLHF self-judge**: model 가 own output 평가.
|
|
|
|
### 매 verify 가 효과적인 곳
|
|
- Multi-hop reasoning (factual chains).
|
|
- Math / logic (intermediate step check).
|
|
- Code (compile, test, lint).
|
|
- Long-form factuality (claim-by-claim).
|
|
- Hallucination 감소.
|
|
|
|
### 매 verify 가 부정확한 곳
|
|
- Model 의 systematic bias — 같은 wrong answer.
|
|
- Highly creative / open-ended (no ground truth).
|
|
- 매 verify model = generator → blind spots 공유.
|
|
|
|
### 매 응용
|
|
1. Agent loop critical-path step 검증.
|
|
2. RAG answer claim verification (cite-check).
|
|
3. Code review pre-PR.
|
|
4. Math homework solver.
|
|
5. Medical / legal high-stakes Q&A.
|
|
|
|
## 💻 패턴
|
|
|
|
### Self-consistency
|
|
```python
|
|
from collections import Counter
|
|
samples = [llm(prompt, temperature=0.8) for _ in range(7)]
|
|
answer = Counter(extract_answer(s) for s in samples).most_common(1)[0][0]
|
|
```
|
|
|
|
### CoVe (4 steps)
|
|
```python
|
|
baseline = llm(f"Answer: {q}")
|
|
verify_qs = llm(f"List 5 verification Qs for: {baseline}")
|
|
verify_as = [llm(f"Answer concisely: {vq}") for vq in verify_qs.splitlines()]
|
|
final = llm(f"Given verification:\n{verify_as}\nRevise: {baseline}")
|
|
```
|
|
|
|
### Self-refine loop
|
|
```python
|
|
draft = llm(f"Solve: {task}")
|
|
for _ in range(3):
|
|
critique = llm(f"Critique:\n{draft}\nList concrete issues; 'NONE' if perfect.")
|
|
if "NONE" in critique[:20]:
|
|
break
|
|
draft = llm(f"Revise based on critique:\n{critique}\n\nDraft:\n{draft}")
|
|
```
|
|
|
|
### Verifier-as-different-model
|
|
```python
|
|
draft = anthropic_call("claude-opus-4-7", task)
|
|
verdict = openai_call("gpt-5", f"Find errors in:\n{draft}")
|
|
final = anthropic_call("claude-opus-4-7", f"Address:\n{verdict}\n\nDraft:\n{draft}")
|
|
```
|
|
|
|
### Code self-test loop
|
|
```python
|
|
code = llm(f"Write Python for: {spec}")
|
|
for _ in range(3):
|
|
res = run_tests(code, spec.tests)
|
|
if res.passed:
|
|
break
|
|
code = llm(f"Tests failed:\n{res.report}\nFix:\n{code}")
|
|
```
|
|
|
|
### Extended thinking (Claude 2026)
|
|
```python
|
|
msg = anthropic.messages.create(
|
|
model="claude-opus-4-7",
|
|
thinking={"type": "enabled", "budget_tokens": 16000},
|
|
messages=[{"role": "user", "content": hard_problem}],
|
|
max_tokens=4096,
|
|
)
|
|
# 매 internal verify already happens within thinking
|
|
```
|
|
|
|
### RAG claim-by-claim verify
|
|
```python
|
|
claims = extract_claims(answer)
|
|
for c in claims:
|
|
evidence = retrieve(c)
|
|
ok = llm(f"Is '{c}' supported by:\n{evidence}\nyes/no")
|
|
if "no" in ok.lower():
|
|
flag(c)
|
|
```
|
|
|
|
## 매 결정 기준
|
|
| 상황 | Approach |
|
|
|---|---|
|
|
| Cheap, parallelizable | self-consistency |
|
|
| Factual long-form | CoVe |
|
|
| Iterative improvement | self-refine |
|
|
| Code / has tests | execution-grounded |
|
|
| Reasoning model 사용 가능 | thinking budget + light verify |
|
|
|
|
**기본값**: thinking + light claim-verify (RAG case) 또는 self-consistency (3-5 samples).
|
|
|
|
## 🔗 Graph
|
|
- 변형: [[Chain-of-Thought]] · [[Self-Consistency]] · [[Reflexion]]
|
|
- 응용: [[RAG]] · [[Code-Generation]]
|
|
|
|
## 🤖 LLM 활용
|
|
**언제**: 매 high-stakes accuracy, hallucination cost 큼. 매 budget 가 latency 보다 중요.
|
|
**언제 X**: 매 latency-critical (chat UI). 매 task 가 verify 가능한 ground truth 없음 (open creative).
|
|
|
|
## ❌ 안티패턴
|
|
- **Self-verify infinite loop**: 매 max iter cap 필수.
|
|
- **Same model verify same model on bias**: blind spots 공유 → cross-model verify.
|
|
- **Verify trivial output**: 매 cost waste — gating 필요.
|
|
- **Trust verify verdict blindly**: verify hallucinate 가능.
|
|
|
|
## 🧪 검증 / 중복
|
|
- Verified (Wang 2022 Self-Consistency, Dhuliawala 2023 CoVe, Madaan 2023 Self-Refine).
|
|
- 신뢰도 A.
|
|
|
|
## 🕓 Changelog
|
|
| 날짜 | 변경 |
|
|
|---|---|
|
|
| 2026-05-08 | Phase 1 |
|
|
| 2026-05-10 | Manual cleanup — verification family + thinking 2026 |
|
|
|
|
## 🛠️ 적용 사례 (Applied in summary)
|
|
|
|
<!-- CODE-GROUNDING:START -->
|
|
### 🔎 코드베이스 근거 (자동 추출 — E:\Wiki 레포)
|
|
**실제 구현/사용 위치:**
|
|
- `connectai/src/features/selfReflector/selfReflectorPrompt.ts:67` — ## [Code Self-Verification — 코드 작성 시 추가 검증]
|
|
|
|
_자동 생성: code_grounding.mjs · 재실행 시 갱신됨_
|
|
<!-- CODE-GROUNDING:END -->
|