The Unified Cognitive Consciousness Theory for Language Models: Anchoring Semantics, Thresholds of Activation, and Emergent Reasoning

TL;DR

This paper introduces a unified theory called UCCT that explains how large language models use external structures to anchor their latent patterns for goal-directed behavior, predicting performance thresholds and generalizing existing learning methods.

Contribution

The paper presents UCCT, a formal model linking anchoring strength to model performance, unifying various in-context learning and tuning techniques under a single theoretical framework.

Findings

Cross-domain anchoring enhances prior knowledge in text and vision.

Ordered thresholds in numeral bases reveal transfer patterns consistent with UCCT.

Layer-wise anchoring correlates with few-shot learning performance.

Abstract

We propose semantic anchoring, a unified account of how large language models turn pretrained capacity into goal-directed behavior: external structure (in-context examples, retrieval, or light tuning) binds the model's latent patterns to desired targets. Unified Contextual Control Theory (UCCT) formalizes this via anchoring strength $S = \rho_d - d_r - \log k$, where $\rho_d$ measures target cohesion in representation space, $d_r$ measures mismatch from prior knowledge, and $k$ is the anchor budget. UCCT predicts threshold-like performance flips and strictly generalizes in-context learning, reading retrieval and fine-tuning as anchoring variants. Three controlled studies provide evidence. Experiment 1 demonstrates cross-domain anchoring rebinding strong priors in text and vision. Experiment 2 varies representational familiarity via numeral bases (base-10/8/9) at fixed complexity, yielding ordered thresholds and transfer patterns tracking $\rho_d$, $d_r$, and $S$. Experiment 3 establishes a geometry-to-behavior correlate: layer-wise peak anchoring and trajectory area predict few-shot thresholds $\theta_{50}$. UCCT offers testable theory and practical metrics for optimizing prompts, retrieval, and tuning.