rust-kgdb 0.6.36 → 0.6.38

package/HYPERMIND_BENCHMARK_REPORT.md

@@ -10,16 +10,19 @@
 
 ## Executive Summary (Verified Results)
 
-**Schema injection + grammar-based predicate resolution improves ALL frameworks by +80.9 percentage points.**
+**Two-stage improvement:**
+1. **Schema injection** brings all frameworks from 0-14% → **71.4%** (+66.7 pp average)
+2. **HyperMind predicate resolver** adds another **+14.3 pp** → **85.7%**
 
-| Framework | No Schema | With Schema | Improvement |
-|-----------|-----------|-------------|-------------|
-| **Vanilla OpenAI** | 0.0% | 85.7% | +85.7 pp |
-| **LangChain** | 0.0% | 85.7% | +85.7 pp |
-| **DSPy** | 14.3% | 85.7% | +71.4 pp |
-| **Average** | 4.8% | **85.7%** | **+80.9 pp** |
+| Framework | No Schema | With Schema | With HyperMind |
+|-----------|-----------|-------------|----------------|
+| **Vanilla OpenAI** | 0.0% | 71.4% | 85.7% |
+| **LangChain** | 0.0% | 71.4% | 85.7% |
+| **DSPy** | 14.3% | 71.4% | 85.7% |
 
-*GPT-4o, 7 LUBM queries, real API calls, no mocking. See `hypermind_improved_benchmark_*.json`.*
+*GPT-4o, 7 LUBM queries (5→6 passed), real API calls, no mocking. See `hypermind_improved_benchmark_*.json`.*
+
+**What the resolver fixed:** Query A1 used "teacher" instead of "teacherOf" - predicate resolver corrected this.
 
 **Key Insight**: The value is in the ARCHITECTURE (schema injection, type contracts), not the specific framework.
 

package/README.md

@@ -57,13 +57,13 @@
 │  MODEL:     GPT-4o │ Real API calls │ No mocking                            │
 ├─────────────────────────────────────────────────────────────────────────────┤
 │                                                                             │
-│  FRAMEWORK         NO SCHEMA     WITH SCHEMA    IMPROVEMENT                 │
+│  FRAMEWORK         NO SCHEMA     WITH SCHEMA    WITH HYPERMIND              │
 │  ─────────────────────────────────────────────────────────────              │
-│  Vanilla OpenAI    0.0%          85.7%          +85.7 pp                    │
-│  LangChain         0.0%          85.7%          +85.7 pp                    │
-│  DSPy              14.3%         85.7%          +71.4 pp                    │
+│  Vanilla OpenAI    0.0%          71.4%          85.7% (+14.3 pp)            │
+│  LangChain         0.0%          71.4%          85.7% (+14.3 pp)            │
+│  DSPy              14.3%         71.4%          85.7% (+14.3 pp)            │
 │  ─────────────────────────────────────────────────────────────              │
-│  AVERAGE           4.8%          85.7%          +80.9 pp                    │
+│  KEY: Schema-aware predicate resolver adds +14.3 pp over schema alone       │
 │                                                                             │
 │  NOTE: Schema injection improves ALL frameworks equally on generation.      │
 │  HyperMind's value = full execution stack, not just generation.             │
@@ -1027,12 +1027,13 @@ console.log('Supersteps:', result.supersteps)  // 5
 
 ### AI Agent Accuracy (Verified December 2025)
 
-| Framework | No Schema | With Schema (HyperMind) | Improvement |
-|-----------|-----------|-------------------------|-------------|
-| **Vanilla OpenAI** | 0.0% | 85.7% | +85.7 pp |
-| **LangChain** | 0.0% | 85.7% | +85.7 pp |
-| **DSPy** | 14.3% | 85.7% | +71.4 pp |
-| **Average** | 4.8% | **85.7%** | **+80.9 pp** |
+| Framework | No Schema | With Schema | With HyperMind |
+|-----------|-----------|-------------|----------------|
+| **Vanilla OpenAI** | 0.0% | 71.4% | 85.7% (+14.3 pp) |
+| **LangChain** | 0.0% | 71.4% | 85.7% (+14.3 pp) |
+| **DSPy** | 14.3% | 71.4% | 85.7% (+14.3 pp) |
+
+*HyperMind's predicate resolver adds +14.3 pp over schema injection alone.*
 
 *Tested: GPT-4o, 7 LUBM queries, real API calls. See `framework_benchmark_*.json` for raw data.*
 
@@ -1451,7 +1452,7 @@ Result: ❌ PARSER ERROR - Invalid SPARQL syntax
 | LangChain | ❌ No | ❌ No | ❌ No | ❌ No |
 | DSPy | ⚠️ Partial | ❌ No | ❌ No | ❌ No |
 
-**Note**: This compares architectural features. Benchmark (Dec 2025): Schema injection improves all frameworks by +80.9 pp (Vanilla: 0%→85.7%, LangChain: 0%→85.7%, DSPy: 14.3%→85.7%).
+**Note**: This compares architectural features. Benchmark (Dec 2025): Schema injection brings all frameworks to 71.4%. HyperMind's predicate resolver adds +14.3 pp to reach 85.7%.
 
 ```
 ┌─────────────────────────────────────────────────────────────────┐

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "rust-kgdb",
-  "version": "0.6.36",
+  "version": "0.6.38",
   "description": "Production-grade Neuro-Symbolic AI Framework with Schema-Aware GraphDB, Context Theory, and Memory Hypergraph: +86.4% accuracy over vanilla LLMs. Features Schema-Aware GraphDB (auto schema extraction), BYOO (Bring Your Own Ontology) for enterprise, cross-agent schema caching, LLM Planner for natural language to typed SPARQL, ProofDAG with Curry-Howard witnesses. High-performance (2.78µs lookups, 35x faster than RDFox). W3C SPARQL 1.1 compliant.",
   "main": "index.js",
   "types": "index.d.ts",