Saved in:
| Main Author: | |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2507.00002 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916817761468416 |
|---|---|
| author | Augeri, Christopher James |
| author_facet | Augeri, Christopher James |
| contents | Large Language Models (LLMs) exhibit remarkable capabilities but suffer from apparent precision loss, reframed here as information spreading. This reframing shifts the problem from computational precision to an information-theoretic communication issue. We address the K:V and V:K memory problem in LLMs by introducing HDRAM (Holographically Defined Random Access Memory), a symbolic memory framework treating transformer latent space as a spread-spectrum channel. Built upon hypertokens, structured symbolic codes integrating classical error-correcting codes (ECC), holographic computing, and quantum-inspired search, HDRAM recovers distributed information through principled despreading. These phase-coherent memory addresses enable efficient key-value operations and Grover-style search in latent space. By combining ECC grammar with compressed sensing and Krylov subspace alignment, HDRAM significantly improves associative retrieval without architectural changes, demonstrating how Classical-Holographic-Quantum-inspired (CHQ) principles can fortify transformer architectures. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2507_00002 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Hypertokens: Holographic Associative Memory in Tokenized LLMs Augeri, Christopher James Machine Learning Artificial Intelligence Computation and Language Large Language Models (LLMs) exhibit remarkable capabilities but suffer from apparent precision loss, reframed here as information spreading. This reframing shifts the problem from computational precision to an information-theoretic communication issue. We address the K:V and V:K memory problem in LLMs by introducing HDRAM (Holographically Defined Random Access Memory), a symbolic memory framework treating transformer latent space as a spread-spectrum channel. Built upon hypertokens, structured symbolic codes integrating classical error-correcting codes (ECC), holographic computing, and quantum-inspired search, HDRAM recovers distributed information through principled despreading. These phase-coherent memory addresses enable efficient key-value operations and Grover-style search in latent space. By combining ECC grammar with compressed sensing and Krylov subspace alignment, HDRAM significantly improves associative retrieval without architectural changes, demonstrating how Classical-Holographic-Quantum-inspired (CHQ) principles can fortify transformer architectures. |
| title | Hypertokens: Holographic Associative Memory in Tokenized LLMs |
| topic | Machine Learning Artificial Intelligence Computation and Language |
| url | https://arxiv.org/abs/2507.00002 |