Cognitive Interception: A Hard Gate Between Attention Scoring and Token Generation
Fuente:
Zenodo
Salvato in:
| Autore principale: | |
|---|---|
| Natura: | Recurso digital |
| Lingua: | inglese |
| Pubblicazione: |
Zenodo
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866901272351735808 |
|---|---|
| author | Dalink, Constant |
| author_facet | Dalink, Constant |
| contents | <pre><code><span>This paper proposes a cognitive interception layer inserted between </span> <span>QK^T attention scoring and token generation in large language models. </span></code></pre> <p> </p> <pre><code><span>The mechanism enforces an "add-then-prune" restructuring of candidate </span> <span>memories within a closed rehearsal space, compresses them into a </span> <span>reflective closure with persistent shadow records for failed paths, </span> <span>and applies a rigid pre-output gate (Check) that separates "can be </span> <span>generated" from "is allowed to be output."</span></code></pre> <p> </p> <pre><code><span>Key contributions:</span></code></pre> <pre><code><span>1. A non-standard algebraic master formula that encodes the full </span> <span> pre-output cognitive lifecycle:</span> <span> ∀± { (V₁,V₂)^c, V₁, V₂, V, … } →[Check] ○ V_use</span></code></pre> <pre><code><span>2. A reflective memory organization mechanism with endogenous </span> <span> classification, random grouping, and structural demotion.</span> </code></pre> <pre><code><span>3. Three verifiable metrics beyond perplexity: semantic redundancy </span> <span> reduction, token compression under equivalent logic, and white-box </span> <span> traceability of failed paths via shadow records.</span></code></pre> <pre><code><span>4. An analysis of Claude 4.6 Opus-Thinking's emergent memory retrieval </span> <span> behavior, arguing that such capabilities can be universalized through </span> <span> structure rather than scale.</span> </code></pre> <pre><code><span>5. Four open riddles on impulse engines, endogenous scoring, random </span> <span> grouping necessity, and motivation perturbation via minimal QK^T </span> <span> deformation.</span></code></pre> <p> </p> <pre><code><span>The paper includes both English and Chinese versions.</span></code></pre> |
| format | Recurso digital |
| id | zenodo_https___doi_org_10_5281_zenodo_19095032 |
| institution | Zenodo |
| language | eng |
| publishDate | 2026 |
| publisher | Zenodo |
| record_format | zenodo |
| spellingShingle | Cognitive Interception: A Hard Gate Between Attention Scoring and Token Generation Dalink, Constant Computer Science Artificial Intelligence Machine Learning Cognitive Science cognitive interception attention mechanism reflective memory organization transformer architecture pre-output gating shadow records GAI alignment interpretability motivation perturbation QKT scoring add-then-prune attention mechanism reflective memory organization transformer architecture pre-output gating shadow records GAI alignment interpretability motivation perturbation QKT scoring <pre><code><span>This paper proposes a cognitive interception layer inserted between </span> <span>QK^T attention scoring and token generation in large language models. </span></code></pre> <p> </p> <pre><code><span>The mechanism enforces an "add-then-prune" restructuring of candidate </span> <span>memories within a closed rehearsal space, compresses them into a </span> <span>reflective closure with persistent shadow records for failed paths, </span> <span>and applies a rigid pre-output gate (Check) that separates "can be </span> <span>generated" from "is allowed to be output."</span></code></pre> <p> </p> <pre><code><span>Key contributions:</span></code></pre> <pre><code><span>1. A non-standard algebraic master formula that encodes the full </span> <span> pre-output cognitive lifecycle:</span> <span> ∀± { (V₁,V₂)^c, V₁, V₂, V, … } →[Check] ○ V_use</span></code></pre> <pre><code><span>2. A reflective memory organization mechanism with endogenous </span> <span> classification, random grouping, and structural demotion.</span> </code></pre> <pre><code><span>3. Three verifiable metrics beyond perplexity: semantic redundancy </span> <span> reduction, token compression under equivalent logic, and white-box </span> <span> traceability of failed paths via shadow records.</span></code></pre> <pre><code><span>4. An analysis of Claude 4.6 Opus-Thinking's emergent memory retrieval </span> <span> behavior, arguing that such capabilities can be universalized through </span> <span> structure rather than scale.</span> </code></pre> <pre><code><span>5. Four open riddles on impulse engines, endogenous scoring, random </span> <span> grouping necessity, and motivation perturbation via minimal QK^T </span> <span> deformation.</span></code></pre> <p> </p> <pre><code><span>The paper includes both English and Chinese versions.</span></code></pre> |
| title | Cognitive Interception: A Hard Gate Between Attention Scoring and Token Generation |
| topic | Computer Science Artificial Intelligence Machine Learning Cognitive Science cognitive interception attention mechanism reflective memory organization transformer architecture pre-output gating shadow records GAI alignment interpretability motivation perturbation QKT scoring add-then-prune attention mechanism reflective memory organization transformer architecture pre-output gating shadow records GAI alignment interpretability motivation perturbation QKT scoring |
| url | https://doi.org/10.5281/zenodo.19095032 |