Paper

Citation

If you use UCEF in your research, please cite:

@article{he2026ucef,
  title     = {UCEF: Universal Context Extension Framework ---
               Breaking the Context Barrier with Hyperbolic Geometry
               and Quantum-Inspired Selection},
  author    = {Honglin He},
  journal   = {arXiv preprint},
  year      = {2026},
  url       = {https://github.com/ViewWay/UCEF}
}

Preprint

• arXiv: Coming soon
• DOI: Pending

Key Contributions

1. Hyperbolic retrieval engine using Poincaré ball embeddings for hierarchical document collections
2. Quantum-inspired context selection via density matrix representation and measurement collapse
3. Adaptive compression with closed-loop quality feedback across 4 dimensions
4. Three-layer memory architecture with automatic document promotion/demotion

Related Publications

Work	Year	Venue	Relation
Poincaré Embeddings	2017	NeurIPS	Foundation for hyperbolic retrieval
HyperbolicRAG	2025	arXiv	Direct predecessor (retrieval only)
HypRAG	2026	arXiv	Hyperbolic dense retrieval for RAG
Quantum-Inspired Self-Attention	2026	arXiv	Quantum attention mechanism
ATACompressor	2026	arXiv	Adaptive task-aware compression