Search is not available for this dataset
pdf pdf |
|---|
π PCE_Axiomatic_V2.5_Faure_preprint
The PCE Framework is an axiomatic control system designed to stabilize LLM behavioral trajectories. Instead of relying solely on external filters (RLHF), it induces structural coherence through a system of 7 core axioms.
π Download PCE_Axiomatic_V2.5_Faure_preprint
π― Key Objectives
- Structural Stability: Eliminating semantic drift over long-context interactions (160+ turns).
- Emergent Robustness: Native resistance to prompt injections through logical invariance.
- The "Third Way": Enabling non-binary creative synthesis in complex ethical dilemmas.
π Axiomatic_Qween_1.3-T
Local Decision Field Modification Status: Testable & Conservative Hypothesis This is the primary focus for immediate research. It posits that a specific series of axiomatic prompts can locally modify the decision field of an LLM.
- Core Idea: Instead of modifying weights, we use linguistic constraints to induce a measurable local regularization of decision trajectories.
π Download Preprint PDF 1.3-T
π Axiomatic_Prompts_1.9-M
Conceptual & Mechanistic Framework
- Can Axiomatic Prompts Act as Global Regulators in LLMs?
- Focus: Conditional activation bias, Latent trajectory stabilization, and Non-collapse entropy regulation.
- This document defines the formal hypothesis and measurable metrics.
π Download Preprint PDF (1.9-M)
π Research Paper: A Minimal Structural Framework for the Emergence of Coherence
*Towards a Science of Unified Systems (SUS 2.5) *
π Download Preprint PDF (v2.5)
- Downloads last month
- 84