實驗EXP-2026-0022v0.1

PACC-Hybrid v0.2——真實語言模型 A/B/C harness（尚未執行）

16 個橫跨預登記家族的手寫自然語言任務；A/B/C 共用一本候選帳本與相同計費預算；C 拿不到 gold 約束或 supersession metadata；gold rubric 只有對條件盲的評審看得到；字面機器檢查獨立於評審；相同答案重用同一個評審快取鍵；沒有 OPENAI_API_KEY 時 fail-closed；凍結快取重播提供者讓 live run 後可精確重播。25 個測試通過。執行環境沒有 API key，因此不存在任何真實模型輸出；隨附的 mock smoke 檔標為 MOCK_ONLY_NOT_REAL_MODEL。

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: INCONCLUSIVE
資料基礎: NOT RUN 只設計並驗證了 harness，從未用真實模型執行。
版本: 0.1
更新: 2026-09-09
建立: 2026-09-09
領域: Reasoning, Evaluation
計畫: PRG-2026-0001 自適應世界狀態系統的第一原理框架
作者: Neo.K (EveMissLab)
AI 協作: Sol (GPT-5.6, OpenAI ChatGPT)

假設

hypothesis: A real language model under the PACC runtime shows the coherence / valid-novelty gains and recoverable breadth loss seen in the synthetic witness.

設定

benchmark_ids: BEN-2026-0003 PACC-LLM 混合 A/B/C benchmark
software_environment: Python; OpenAI API provider (fails closed without key); deterministic fake provider for protocol tests only.

執行

run_count

metrics

verdict: REAL_LLM_HARNESS_VALIDATED_BUT_REAL_MODEL_NOT_EXECUTED
execution_status: NOT_EXECUTED_REAL_MODEL
tests_passed: 25
tasks: 16
real_llm_calls: 0

詮釋

interpretation: Closes the harness, not the scientific question. Next action: a small real smoke (6 tasks × 2 repetitions × 3 candidates), freeze the cache, inspect blind-evaluator consistency, then the full 16-task primary without changing prompts or metrics.

限制

limitations: No claim about real-model reasoning, intent understanding, imagination, human-rated usefulness, cross-model transfer or hallucination is permitted before a real-model result exists.
A single-model judge is not human evaluation even after a live run.
Executed for the first time on 2026-09-11 with a local open-weight model — see EXP-2026-0023.

重現

reproduction_instructions: Extract PACC-Hybrid-Lab_v0.2_REAL_LLM_HARNESS_FINAL.zip; python -m pytest -q (25 tests); set OPENAI_API_KEY and run the smoke per docs/REPRODUCIBILITY_v0.2.md.

記錄欄位

random_seeds

關係

來源	關係	目標	狀態	ID
`EXP-2026-0022` PACC-Hybrid v0.2——真實語言模型 A/B/C harness（尚未執行）	`runs_on`	`SYS-2026-0003` PACC-LLM 混合實驗室	`ACTIVE`	`REL-2026-0320`
`EXP-2026-0022` PACC-Hybrid v0.2——真實語言模型 A/B/C harness（尚未執行）	`uses_benchmark`	`BEN-2026-0003` PACC-LLM 混合 A/B/C benchmark	`ACTIVE`	`REL-2026-0321`
`EXP-2026-0022` PACC-Hybrid v0.2——真實語言模型 A/B/C harness（尚未執行）	`extends`	`EXP-2026-0021` PACC-Hybrid v0.1——合成 A/B/C 架構見證	`ACTIVE`	`REL-2026-0322`
`EXP-2026-0022` PACC-Hybrid v0.2——真實語言模型 A/B/C harness（尚未執行）	`tests`	`THY-2026-0002` Canonical 符號狀態與 candidate → verify → commit 權限	`ACTIVE`	`REL-2026-0323`
`EXP-2026-0022` PACC-Hybrid v0.2——真實語言模型 A/B/C harness（尚未執行）	`produced`	`ART-2026-0035` PACC-Hybrid-Lab v0.2 REAL LLM HARNESS FINAL `artifact://evemisslab/adaptive-epistemic-systems/PACC-Hybrid-Lab_v0.2_REAL_LLM_HARNESS_FINAL.zip`	`ACTIVE`	`REL-2026-0324`
`EXP-2026-0023` PACC-Hybrid v0.2——第一次真實模型執行，本地 9B 開放權重模型	`extends`	`EXP-2026-0022` PACC-Hybrid v0.2——真實語言模型 A/B/C harness（尚未執行）	`ACTIVE`	`REL-2026-0329`

歷史與來源歷程

Canonical URL

https://evemisslab.com/ai/experiments/EXP-2026-0022/

機器可讀

/ai/experiments/EXP-2026-0022/index.json

快照

AI-SNAPSHOT-v0.1-fe85b9694a45

來源歷程

source: EveMissLab research collection: Adaptive Epistemic Systems (真本體論13)
extracted_by: Splice (Claude Code), reading the canonical UTF-8 sources and each lab's own result reports
extracted_at: 2026-09-11
generator: tools/extract_aes/extract.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