feat: complete wikification of War Commander batch 1&2 and final grey dot cleanup

10_Wiki/Topics/AI/Best-of-N-Sampling.md

@@ -1,6 +1,6 @@
 ---
 id: P-REINFORCE-AUTO-BONS-001
-category: "[[10_Wiki/💡 Topics/AI]]"
+category: "10_Wiki/💡 Topics/AI"
 confidence_score: 0.94
 tags: [auto-reinforced, best-of-n, sampling-strategy, inference-optimization, llm, reasoning, reranking]
 last_reinforced: 2026-04-20
@@ -26,6 +26,6 @@ Best-of-N Sampling(최적 샘플링)은 거대 언어 모델(LLM)의 추론 품
 - **정책 변화(RL Update)**: 최근 OpenAI o1 등 추론 전문 모델 정책은 단순히 N개를 뽑는 수준을 넘어, 생각의 체인(CoT) 과정 자체를 검증하고 수정하는 시스템으로 진화 중임. (Tree-of-Thought와 연결)
 
 ## 🔗 지식 연결 (Graph)
-- [[Scalability]], [[Reinforcement Learning (RL)]], [[Tree-of-Thought]], [[Search-Strategy]], [[Inference]]
+- [[Scalability]], [[Reinforcement Learning (RL)]], Tree-of-Thought, [[Search-Strategy]], Inference
 - **Related Terms**: Rejection Sampling, Majority Voting, Thought-level Verifiers.
 ---