2nd/10_Wiki/Topics/AI_and_ML/Reflection.md

id, title, category, status, canonical_id, aliases, duplicate_of, source_trust_level, confidence_score, verification_status, tags, raw_sources, last_reinforced, github_commit, tech_stack

id	title	category	status	canonical_id	aliases	duplicate_of	source_trust_level	confidence_score	verification_status	tags	raw_sources	last_reinforced	github_commit	tech_stack
wiki-2026-0508-reflection	Reflection	10_Wiki/Topics	verified	self	Self-Reflection Reflexion Programming Reflection	none	A	0.9	applied	reflection metaprogramming llm reflexion self-critique		2026-05-10	pending	language framework python anthropic-sdk

Reflection

매 한 줄

"매 program inspects itself, agent critiques itself". Reflection 은 dual concept — programming 에서 runtime 의 type/method introspection, AI 에서 LLM 의 self-critique loop. 2023 Reflexion paper (Shinn) 가 후자를 popularize, 2026 의 agent loop 의 backbone.

매 핵심

매 Programming Reflection

• Java: Class.forName, Method.invoke — runtime type lookup.
• Python: getattr, inspect, type() — first-class.
• Go: reflect.ValueOf, reflect.TypeOf — verbose but explicit.
• Rust: 매 limited — std::any::Any, no runtime method dispatch.
• Cost: 매 10-100x slower than direct call. JIT 의 mitigates partially.

매 LLM Reflection

• Reflexion (2023): agent generates → critiques → retries with verbal feedback in context.
• Self-critique: 매 model evaluates own output against rubric/spec.
• Constitutional AI: Anthropic 의 자기-revision against principles.
• CRITIC (2024): tool-augmented self-correction.
• Test-time compute (o1, Claude thinking): 매 internal reflection 의 productized.

매 응용

1. Agent error recovery — failed tool call 의 self-diagnose.
2. Code generation — write → test → fix loop.
3. Math/logic — chain-of-thought + verifier.
4. Plugin systems — runtime method discovery.
5. ORM — entity-to-table reflection mapping.

💻 패턴

Python introspection

import inspect

class Service:
    def fetch(self, url: str) -> dict: ...
    def post(self, url: str, body: dict) -> None: ...

svc = Service()
for name, method in inspect.getmembers(svc, predicate=inspect.ismethod):
    sig = inspect.signature(method)
    print(f"{name}{sig}")

Go reflect (struct tags)

type User struct {
    Name  string `json:"name" validate:"required"`
    Email string `json:"email" validate:"email"`
}

func validate(v any) error {
    t := reflect.TypeOf(v)
    val := reflect.ValueOf(v)
    for i := 0; i < t.NumField(); i++ {
        tag := t.Field(i).Tag.Get("validate")
        if tag == "required" && val.Field(i).IsZero() {
            return fmt.Errorf("%s required", t.Field(i).Name)
        }
    }
    return nil
}

Reflexion loop (Claude Opus 4.7)

from anthropic import Anthropic

client = Anthropic()
MODEL = "claude-opus-4-7"

def reflexion(task: str, max_iter: int = 3) -> str:
    history = []
    attempt = generate(task, history)
    for i in range(max_iter):
        critique = client.messages.create(
            model=MODEL,
            max_tokens=1024,
            messages=[{
                "role": "user",
                "content": f"Task: {task}\nAttempt: {attempt}\n"
                           f"Critique: list errors, missing edge cases. "
                           f"If perfect, reply DONE."
            }]
        ).content[0].text
        if "DONE" in critique:
            return attempt
        history.append({"attempt": attempt, "critique": critique})
        attempt = generate(task, history)
    return attempt

def generate(task: str, history: list) -> str:
    ctx = "\n".join(f"Prev attempt: {h['attempt']}\nCritique: {h['critique']}"
                    for h in history)
    resp = client.messages.create(
        model=MODEL,
        max_tokens=2048,
        messages=[{"role": "user", "content": f"{ctx}\n\nTask: {task}"}]
    )
    return resp.content[0].text

Self-critique with extended thinking

# Claude Opus 4.7 의 native thinking — implicit reflection
resp = client.messages.create(
    model="claude-opus-4-7",
    max_tokens=8192,
    thinking={"type": "enabled", "budget_tokens": 4096},
    messages=[{"role": "user", "content": "Solve: ..."}]
)
# resp.content[0].type == "thinking" — reflection trace
# resp.content[1].type == "text" — final answer

Plugin discovery (Java)

ServiceLoader<Plugin> loader = ServiceLoader.load(Plugin.class);
for (Plugin p : loader) {
    Method init = p.getClass().getMethod("init", Config.class);
    init.invoke(p, cfg);
}

매 결정 기준

상황	Approach
Hot loop performance	매 reflection X — codegen / direct call
Plugin / DI framework	Reflection OK (one-time init)
LLM agent error recovery	Reflexion loop (max 3 iter)
Math/code with verifier	Self-critique + tool execution
Chat UX	Extended thinking (native)

기본값: 매 native thinking (Claude Opus 4.7) 의 first try, explicit Reflexion 의 verifiable domain (code/math).

🔗 Graph

• 부모: Metaprogramming
• 변형: Reflexion · AI_Safety_and_Alignment · Self-Consistency
• 응용: Code-Generation · Tool-Use
• Adjacent: Chain-of-Thought

🤖 LLM 활용

언제: 매 verifiable output (code passes tests, math 의 numerical), expensive failures (production agent), multi-step planning. 언제 X: simple Q&A (reflection 의 cost > benefit), creative generation (critique 의 collapse to mean), latency-critical (<500ms).

❌ 안티패턴

• Infinite reflection: 매 max_iter cap 없음 → cost runaway.
• Critic = generator: same model 의 critique 의 weak — 매 stronger verifier 의 use.
• Reflection in hot path: Java/Go 의 production code 의 reflect.Value 의 hide.
• Verbal-only critique: 매 numeric/test signal 없음 → noise.
• Self-praise loop: critique prompt 의 "find ANY issue" 의 biased.

🧪 검증 / 중복

• Verified (Reflexion paper 2023, Anthropic Constitutional AI, Java/Go reflect docs).
• 신뢰도 A.

🕓 Changelog

날짜	변경
2026-05-08	Phase 1
2026-05-10	Manual cleanup — programming + LLM reflection unified