研究RES-2026-0004v0.1
PACC 式 runtime 用在語言模型上：推理、意圖與創造廣度
把候選選擇包進 canonical 硬約束／衍生約束的 commit 空間，會改變生成器的產出嗎？合成的 A/B/C 見證（純生成器／硬驗證器／PACC runtime）顯示衍生一致性與滿足約束的新穎度上升、軟偏好契合略降，以及一個可由事後「elastic」探索策略恢復的創造廣度塌縮。真實語言模型版本的 benchmark 已有驗證過、fail-closed 的 harness，但尚未執行。
研究問題
research_questions- Is there a reasoning-up / imagination-down trade-off, or is the observed breadth loss a selection-policy artefact separable from the commit constraints?
- Will a frontier language model show the same effect as the synthetic witness?
主張
claims- Within the synthetic witness: reasoning coherence ↑, valid novelty ↑, soft-preference fit slightly ↓, creative breadth ↓ under naive selection and recoverable with exploration separated from commitment (8/8 seeds sign-stable).
- Real local 9B model, thinking off, two runs (32 and 64 rows per condition): the three runtime conditions are practically equivalent on every judge axis, and by label-free measures the PACC runtime does not narrow creative breadth. The synthetic v0.1 prediction did not appear on this model.
- Real local 9B model with thinking enabled (32 rows per condition): the PACC runtime has the highest valid novelty (+0.069 vs the hard verifier, +0.038 vs the plain model) and repair, governance axes are flat to slightly lower, and breadth is again not reduced — half of the synthetic prediction, unreplicated.
限制
limitations- Zero real LLM calls so far; no claim about real-model reasoning, intent understanding, imagination, hallucination or human-rated usefulness is permitted before a real-model result exists.
- The v0.1 numbers are synthetic data and theoretical reasoning; until a real hybrid model exists, an inference is only an inference — theoretically possible is not actually possible.
關係
| 來源 | 關係 | 目標 | 狀態 | ID |
|---|---|---|---|---|
RES-2026-0004 PACC 式 runtime 用在語言模型上：推理、意圖與創造廣度 | belongs_to | PRG-2026-0001 自適應世界狀態系統的第一原理框架 | ACTIVE | REL-2026-0004 |
RES-2026-0004 PACC 式 runtime 用在語言模型上：推理、意圖與創造廣度 | extends | RES-2026-0002 PACC 猜想——非概率 primitive 會不會收斂到概率表象？ | ACTIVE | REL-2026-0006 |
SYS-2026-0003 PACC-LLM 混合實驗室 | supports | RES-2026-0004 PACC 式 runtime 用在語言模型上：推理、意圖與創造廣度 | ACTIVE | REL-2026-0087 |
DAT-2026-0002 PACC-Hybrid v0.1 共享候選池 | supports | RES-2026-0004 PACC 式 runtime 用在語言模型上：推理、意圖與創造廣度 | ACTIVE | REL-2026-0093 |
歷史與來源歷程
- Canonical URL
- https://evemisslab.com/ai/research/RES-2026-0004/
- 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