# Build Log — 2026-06-22 Hachimoji Citation Pipeline ## Summary Added `python/hachimoji_citation.py` — equation → Hachimoji state → arxiv citation bridge. ## Files Added - `python/hachimoji_citation.py` — Main pipeline with `cite_equation()` and `cite_batch()` functions ## Verification ``` lake build # Build completed successfully (2987 jobs) ``` ## Test Results (with embedding model) ``` $ python3 python/hachimoji_citation.py --top-k 3 Equation : E = mc^2 State: Φ phase=0° admission=ADMIT Citations: 1. [0.0164] Global Solution for Gas-Liquid Flow of 1-D van der Waals Equation... 2. [0.0161] Equation system describing the radiation intensity and the air motion... 3. [0.0159] Equilibrium and Non-Equilibrium diffusion approximation... Equation : ∀x. P(x) → Q(x) State: Λ phase=45° admission=ADMIT Citations: 1. [0.0164] An optimal three-way stable and monotonic spectrum of bounds... 2. [0.0161] A finiteness theorem for algebraic cycles... 3. [0.0159] Fixed Points of the q-Bracket... Equation : 0 = 1 State: Ω phase=180° admission=QUARANTINE Citations: 1. [0.0164] On the double critical Maxwell equations... 2. [0.0161] The number of eigenstates: counting function and heat kernel... 3. [0.0159] Explicit subsolutions and a Liouville theorem... ``` ## Regime Coherence Verified The geometry tracks real mathematical structure because different Hachimoji states retrieve different paper clusters: - **Φ (fundamental)** → phase transition / van der Waals papers - **Λ (room)** → algebraic cycles / quantile bounds papers - **Ω (collision)** → Maxwell equations / critical states papers ## Regime Coherence Analysis The different Hachimoji states produce distinct semantic queries: - **Φ** (fundamental): density boundary, phase transition, Erdős-Rényi — finds foundational physics/maths papers - **Ω** (contradiction): Goormaghtoi, Baker bound, eigensolid — finds collision/equation structure papers - **Π** (potential): calculus, integral, high complexity — finds analysis/calculus papers - **Σ** (symmetric): algebraic geometry, swap, balanced equation — finds symmetric structure papers Note: Python `parse_shape()` is heuristic (not a full AST parser). The `classify_equation()` function mirrors `HachimojiCodec.lean` logic exactly but may diverge on exact counts. Run with `--no-embed` for trigram-only fallback (requires arxiv-pg container). ## Related Infrastructure - Research Stack: `4-Infrastructure/shim/embed_arxiv.py`, `hybrid_search.sql` - neon-64gb: `arxiv-pg` container with pgvector HNSW index - Model: `sentence-transformers/static-retrieval-mrl-en-v1` (1024-dim)