實驗EXP-2026-0102v0.1

XA-05——以腳本化供應商跑的 36 試驗端到端煙霧閘（合成）

用 ScriptedProvider 把 MATH-003、CODE-001、CON-003 × A0–A5 × 2 次跑過 XA-04 + XA-03 + XA-02，輸出被刻意設計成品質曲線事先已知（A0 0.30、A1 0.633、A2–A5 1.0 → SSR 0.30／SDR 0.70 純屬夾具）。36 次試驗、180 次呼叫、168 條軌跡、6 次重試、6 次工具呼叫、24 次驗證；候選 168 = 36 選中 + 132 丟棄；會計不符 0 件、私有參考洩漏 0 件；能量刻意為 null（NullCollector）以證明 Unknown ≠ 0。套件明寫 scientific_interpretation_allowed: false。

研究狀態: STABLE 目前的研究結論相對穩定
證據等級: E1 內部觀察
結果: POSITIVE
資料基礎: SYNTHETIC 合成數據與理論推理。現在很多人把合成數據當成真的；這個實驗室刻意反過來說——在真正的混合模型出現之前，推論就只是推論，理論上可能不等於實際上可能。
版本: 0.1
更新: 2026-09-03
建立: 2026-09-03
領域: Evaluation, Agent Systems
計畫: PRG-2026-0101 智能的物理計量（IPM）
作者: Neo.K (EveMissLab)
AI 協作: Aletheia (GPT-5.6 Sol, OpenAI ChatGPT)

假設

hypothesis: Engineering only: the XA-01→XA-04 pipeline reconstructs a pre-designed A0→A5 quality structure with exact candidate accounting and no private-reference leakage.

設定

benchmark_ids: BEN-2026-0101 XA-02——A0→A5 鷹架響應的 30 題 pilot 任務包
software_environment: XA-05 v0.1 over XA-02/03/04 (hashes declared in README and asserted by this extractor)

程序

procedure: python -m pytest under IPM_XA02_ROOT / IPM_XA03_ROOT / IPM_XA04_ROOT; verify_canonical_output('canonical_smoke_run') re-validates the packaged run from the extracted package.

執行

run_count

controls

scripted outputs with a known oracle
NullCollector so no physical number can masquerade as measurement
dependency packages pinned by SHA-256

metrics

gate

trial_count: 36
all_trials_complete: true
accounting_mismatches: 0
private_sentinel_leaks: 0
ssr: 0.3
sdr: 0.7
energy_scm
scientific_interpretation_allowed: false

execution_totals

created: 168
discarded: 132
model: 180
retry: 6
selected: 36
tool: 6
trajectory: 168
verifier: 24

詮釋

interpretation: The instrument works as an accounting machine: candidate identity 168 = 36 + 132 holds, nothing leaks, the designed curve comes back out. Nothing here is a statement about any model — the package forbids reading it that way, and this record carries the SYNTHETIC badge for the same reason.

限制

limitations: Synthetic fixtures; three tasks; no physical telemetry by design.

重現

reproduction_instructions: Extract XA-02, XA-03, XA-04 and XA-05 as siblings, export the three roots, PYTHONPATH=. python -m pytest -q; or verify the shipped canonical_smoke_run/.

關係

來源	關係	目標	狀態	ID
`EXP-2026-0102` XA-05——以腳本化供應商跑的 36 試驗端到端煙霧閘（合成）	`extends`	`EXP-2026-0101` Experiment A——單次智能與鷹架增益的受控實驗協定 v0.1	`ACTIVE`	`REL-2026-0466`
`EXP-2026-0102` XA-05——以腳本化供應商跑的 36 試驗端到端煙霧閘（合成）	`uses_benchmark`	`BEN-2026-0101` XA-02——A0→A5 鷹架響應的 30 題 pilot 任務包	`ACTIVE`	`REL-2026-0467`
`EXP-2026-0102` XA-05——以腳本化供應商跑的 36 試驗端到端煙霧閘（合成）	`runs_on`	`SYS-2026-0101` XA-03——遙測與執行記錄器（物理執行證據層）	`ACTIVE`	`REL-2026-0468`
`EXP-2026-0102` XA-05——以腳本化供應商跑的 36 試驗端到端煙霧閘（合成）	`runs_on`	`SYS-2026-0102` XA-04——A0→A5 模型執行器與鷹架調度器	`ACTIVE`	`REL-2026-0469`
`EXP-2026-0102` XA-05——以腳本化供應商跑的 36 試驗端到端煙霧閘（合成）	`produced`	`ART-2026-0116` EML-IPM-XA-05 v0.1 — 36-trial end-to-end synthetic smoke gate with canonical scripted run `artifact://evemisslab/intelligence-physical-metrology/IPM_v0.2_XA05_36Trial_SmokeGate_v0.1.zip`	`ACTIVE`	`REL-2026-0470`
`EXP-2026-0103` XA-06 第一次真實模型 pilot——Qwythos-9B-v2 走 A0→A5 階梯（36 試驗，2026-09-03）	`extends`	`EXP-2026-0102` XA-05——以腳本化供應商跑的 36 試驗端到端煙霧閘（合成）	`ACTIVE`	`REL-2026-0472`

歷史與來源歷程

Canonical URL

https://evemisslab.com/ai/experiments/EXP-2026-0102/

機器可讀

/ai/experiments/EXP-2026-0102/index.json

快照

AI-SNAPSHOT-v0.1-fe85b9694a45

來源歷程

source: EveMissLab research collection: Intelligence Physical Metrology (真本體論13)
extracted_by: Splice (Claude Code), reading the canonical UTF-8 sources and each package's own reports
extracted_at: 2026-09-11
generator: tools/extract_all.py
claim_boundary: status, evidence level and result type follow the source artifact's own stated claim boundary; nothing is upgraded beyond what the report supports