The Dabbawala Codex: Hierarchical Bitmask Tries for Zero-Hallucination AI Knowledge Routing
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
The paper introduces the Hierarchical Bitmask Trie (HBT) — a 5-level tree data structure where each level carries its own bitmask, enabling sub-millisecond traversal from query intent to matched rule with zero string parsing and zero hallucination. Named after the Mumbai dabbawala delivery system which achieves 99.9999% accuracy using a 6-character alphanumeric routing label — itself a form of hierarchical bitmask — the HBT extends the proven 32-bit `trust_mask` architecture to cover rule authorship (Kitchen Level), domain branches, service nodes, and individual rule leaves. The novel Kitchen Level bitmask makes rule provenance machine-verifiable for the first time. The PHALA return signal (from the PRAMANA cognition protocol) routes delivery outcomes directly to the Kitchen Level — closing the authorship loop. ANKR (PowerPbox IT Solutions) is the reference implementation with 208 services, 5,336 rules, and the HBT as the unified capability and knowledge routing layer.
Fields of Study & Subjects
Algorithms and Data Compression
Advanced Database Systems and Queries
Network Packet Processing and Optimization
Publication details
- Journal / source
- Zenodo (CERN European Organization for Nuclear Research)
- Publication date
- 28 March 2026
- Language
- EN
- Document type
- Preprint