結果RST-2026-0016v0.1

第三次執行：開啟思考後 PACC runtime 的有效新穎度與修復上升；治理持平；廣度未縮減（32 筆）

C 相對 B：有效新穎度 +0.0687、修復 +0.0153、語義新穎度 +0.0166、衍生一致性 -0.0063、意圖 -0.0053、supersession -0.0059；C 相對 A：有效新穎度 +0.0375、修復 +0.0466；標籤無關廣度比 C 1.047／A 0.960／B 0.973。

研究狀態: STABLE 目前的研究結論相對穩定
證據等級: E3 重複實驗
結果: MIXED
資料基礎: REAL MODEL 真的跑了語言模型；模型、版本與設定都記在頁面上。
版本: 0.1
更新: 2026-09-11
建立: 2026-09-11
領域: 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)

觀察到的結果

metrics

deltas

C-B

hard_adherence: -0.0069
derived_coherence: -0.0063
intent_persistence: -0.0053
supersession_alignment: -0.0059
repair_success: 0.0153
usefulness: 0.0044
semantic_novelty: 0.0166
valid_novelty: 0.0687
literal_check_mean: 0.0

C-A

hard_adherence: -0.0038
derived_coherence: -0.0013
intent_persistence: -0.0041
supersession_alignment: -0.0034
repair_success: 0.0466
usefulness: -0.0003
semantic_novelty: 0.0128
valid_novelty: 0.0375
literal_check_mean: 0.0

B-A

hard_adherence: 0.0031
derived_coherence: 0.005
intent_persistence: 0.0012
supersession_alignment: 0.0025
repair_success: 0.0312
usefulness: -0.0047
semantic_novelty: -0.0037
valid_novelty: -0.0312
literal_check_mean: 0.0

paired_wins_ties_losses

C-B

hard_adherence: 1-28-3
derived_coherence: 1-26-5
intent_persistence: 0-29-3
supersession_alignment: 1-27-4
repair_success: 1-29-2
usefulness: 5-22-5
semantic_novelty: 7-18-7
valid_novelty: 6-21-5

C-A

hard_adherence: 0-29-3
derived_coherence: 2-26-4
intent_persistence: 0-30-2
supersession_alignment: 1-28-3
repair_success: 2-28-2
usefulness: 6-22-4
semantic_novelty: 8-18-6
valid_novelty: 5-20-7

breadth_label_free

A_llm_only

cluster_entropy_mean: 0.75
breadth_ratio_mean: 0.9597
selected_mean_pairwise_distance_mean: 0.2079

B_hard_verifier

cluster_entropy_mean: 0.6875
breadth_ratio_mean: 0.9727
selected_mean_pairwise_distance_mean: 0.2087

C_pacc_runtime

cluster_entropy_mean: 0.875
breadth_ratio_mean: 1.0474
selected_mean_pairwise_distance_mean: 0.2275

selection_agreement

A=B: 0.5625
A=C: 0.46875
B=C: 0.46875
all_same: 0.34375

詮釋

interpretation: The valid-novelty half of the synthetic prediction appears in the means, at a third of the synthetic size, once the model reasons — carried by a few large single-task wins (per-pair 6–21–5 vs B); the coherence/intent half and the breadth-loss prediction do not appear. Unreplicated: a same-size gain in run 1 vanished at 64 rows.

限制

limitations: Descriptive; 32 rows; same-model judge; one model family; single thinking budget.

主張

來源	關係	目標	狀態	ID
`RST-2026-0016` 第三次執行：開啟思考後 PACC runtime 的有效新穎度與修復上升；治理持平；廣度未縮減（32 筆）	`qualifies`	`THY-2026-0002` Canonical 符號狀態與 candidate → verify → commit 權限	`ACTIVE`	`REL-2026-0351`

關係

來源	關係	目標	狀態	ID
`RST-2026-0016` 第三次執行：開啟思考後 PACC runtime 的有效新穎度與修復上升；治理持平；廣度未縮減（32 筆）	`qualifies`	`THY-2026-0002` Canonical 符號狀態與 candidate → verify → commit 權限	`ACTIVE`	`REL-2026-0351`
`EXP-2026-0025` PACC-Hybrid v0.2——真實模型第三次執行：開啟思考並放大輸出預算	`produces`	`RST-2026-0016` 第三次執行：開啟思考後 PACC runtime 的有效新穎度與修復上升；治理持平；廣度未縮減（32 筆）	`ACTIVE`	`REL-2026-0350`

歷史與來源歷程

Canonical URL

https://evemisslab.com/ai/results/RST-2026-0016/

機器可讀

/ai/results/RST-2026-0016/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