Spaces:
Running
Running
Publish VaidhyaMegha Organization Card (Samyama Graph, 19 KGs, Shoonya LLM)
Browse files
README.md
CHANGED
|
@@ -1,10 +1,97 @@
|
|
| 1 |
---
|
| 2 |
-
title:
|
| 3 |
-
emoji:
|
| 4 |
colorFrom: indigo
|
| 5 |
-
colorTo:
|
| 6 |
sdk: static
|
| 7 |
-
pinned:
|
| 8 |
---
|
| 9 |
|
| 10 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
+
title: VaidhyaMegha — Open Graph Systems, Knowledge Graphs & First-Principles Language Models
|
| 3 |
+
emoji: 🧭
|
| 4 |
colorFrom: indigo
|
| 5 |
+
colorTo: blue
|
| 6 |
sdk: static
|
| 7 |
+
pinned: true
|
| 8 |
---
|
| 9 |
|
| 10 |
+
# VaidhyaMegha & Samyama Research
|
| 11 |
+
|
| 12 |
+
**VaidhyaMegha Private Limited** builds first-principles systems at the intersection of **high-performance graph databases (`Samyama Graph`)**, **curated multi-domain Knowledge Graphs (`19` open KG datasets)**, and **mechanistically instrumented language models (`Shoonya`)**.
|
| 13 |
+
|
| 14 |
+
Our engineering philosophy across every initiative is simple: **zero black boxes, Rust-native systems rigor, mathematical & spectral observability, and reproducible open benchmarks.**
|
| 15 |
+
|
| 16 |
+
---
|
| 17 |
+
|
| 18 |
+
## 1. The Samyama Graph Initiative (`samyama-graph`)
|
| 19 |
+
|
| 20 |
+
**Samyama Graph** is a Rust-native, multi-tenant property graph, vector (`HNSW`), and Cypher query engine speaking the Redis wire protocol (`RESP`) alongside HTTP and Model Context Protocol (`MCP`) endpoints for zero-hallucination `GraphRAG` and agentic reasoning.
|
| 21 |
+
|
| 22 |
+
| Benchmark Suite | Verified Result | Scale / Evaluation Scope |
|
| 23 |
+
| :--- | :--- | :--- |
|
| 24 |
+
| **LDBC SNB Interactive** | **21 / 21 (100%)** | Scale Factor 1 (`3.18M` nodes, `17.26M` edges) |
|
| 25 |
+
| **LDBC Graphalytics** | **12 / 12 (100%)** | Reference graph algorithm verification |
|
| 26 |
+
| **LDBC FinBench** | **40 / 40 (100%)** | Financial transaction & fraud pattern queries |
|
| 27 |
+
| **LDBC SNB Business Intelligence** | **16 / 16 executed** | Complex analytical OLAP graph aggregations |
|
| 28 |
+
| **BiomedQA (Federated Clinical KGs)** | **98% (MCP) / 85% (NLQ)** | vs. `75%` ungrounded frontier LLM baseline across 3 federated KGs |
|
| 29 |
+
| **AssetOpsBench (Industrial Operations)** | **99% (MCP) / 83% (NLQ)** | vs. `65%` ungrounded LLM baseline across 139 industrial scenarios |
|
| 30 |
+
|
| 31 |
+
---
|
| 32 |
+
|
| 33 |
+
## 2. Open Knowledge Graph Datasets (`19` Curated KGs on Hugging Face)
|
| 34 |
+
|
| 35 |
+
To advance reproducible **GraphRAG**, **Knowledge-Grounded Reasoning**, **Graph-Aware Embeddings**, and **Multi-Hop Structured QA**, VaidhyaMegha publishes **19 domain-specific Knowledge Graph datasets** spanning scientific literature, clinical medicine, financial risk, legal reasoning, industrial operations, and sports analytics:
|
| 36 |
+
|
| 37 |
+
### Scientific & Systems Research Knowledge Graphs
|
| 38 |
+
| Dataset | Domain Focus |
|
| 39 |
+
| :--- | :--- |
|
| 40 |
+
| [`VaidhyaMegha/ai-research-kg`](https://huggingface.co/datasets/VaidhyaMegha/ai-research-kg) | AI/ML architectures, benchmarks, citations, methods, and lineage |
|
| 41 |
+
| [`VaidhyaMegha/maths-research-kg`](https://huggingface.co/datasets/VaidhyaMegha/maths-research-kg) | Mathematical theorems, structures, proof dependencies, and conjectures |
|
| 42 |
+
| [`VaidhyaMegha/dbms-research-kg`](https://huggingface.co/datasets/VaidhyaMegha/dbms-research-kg) | Database internals, query optimizers, storage engines, and concurrency protocols |
|
| 43 |
+
|
| 44 |
+
### Healthcare, Life Sciences & Regulatory Knowledge Graphs
|
| 45 |
+
| Dataset | Domain Focus |
|
| 46 |
+
| :--- | :--- |
|
| 47 |
+
| [`VaidhyaMegha/clinicaltrials-kg`](https://huggingface.co/datasets/VaidhyaMegha/clinicaltrials-kg) | Clinical trial protocols, interventions, eligibility criteria, and outcomes |
|
| 48 |
+
| [`VaidhyaMegha/pathways-kg`](https://huggingface.co/datasets/VaidhyaMegha/pathways-kg) | Biological & molecular signaling pathways, gene-protein interactions |
|
| 49 |
+
| [`VaidhyaMegha/health-determinants-kg`](https://huggingface.co/datasets/VaidhyaMegha/health-determinants-kg) | Social, environmental, and population-level determinants of health |
|
| 50 |
+
| [`VaidhyaMegha/mental-health-kg`](https://huggingface.co/datasets/VaidhyaMegha/mental-health-kg) | Psychiatric conditions, diagnostic criteria, interventions, and care pathways |
|
| 51 |
+
| [`VaidhyaMegha/regulatory-affairs-kg`](https://huggingface.co/datasets/VaidhyaMegha/regulatory-affairs-kg) | Medical & pharmaceutical regulatory frameworks, submissions, and compliance |
|
| 52 |
+
|
| 53 |
+
### Banking, Finance, Model Risk & Legal Knowledge Graphs
|
| 54 |
+
| Dataset | Domain Focus |
|
| 55 |
+
| :--- | :--- |
|
| 56 |
+
| [`VaidhyaMegha/banking-kg`](https://huggingface.co/datasets/VaidhyaMegha/banking-kg) | Retail & commercial banking entities, products, transactions, and controls |
|
| 57 |
+
| [`VaidhyaMegha/finance-kg`](https://huggingface.co/datasets/VaidhyaMegha/finance-kg) | Financial instruments, market structures, corporate filings, and indicators |
|
| 58 |
+
| [`VaidhyaMegha/bank-model-risk-kg`](https://huggingface.co/datasets/VaidhyaMegha/bank-model-risk-kg) | SR 11-7 model risk management (`MRM`), validation findings, and governance |
|
| 59 |
+
| [`VaidhyaMegha/legal-judgments-kg`](https://huggingface.co/datasets/VaidhyaMegha/legal-judgments-kg) | Judicial precedents, statutes, citations, holdings, and legal issues |
|
| 60 |
+
|
| 61 |
+
### Industrial, Edge AI, Civic & EdTech Knowledge Graphs
|
| 62 |
+
| Dataset | Domain Focus |
|
| 63 |
+
| :--- | :--- |
|
| 64 |
+
| [`VaidhyaMegha/assetops-kg`](https://huggingface.co/datasets/VaidhyaMegha/assetops-kg) | Industrial asset hierarchy, sensor telemetry, failure modes (`FMEA`), and maintenance |
|
| 65 |
+
| [`VaidhyaMegha/devops-kg`](https://huggingface.co/datasets/VaidhyaMegha/devops-kg) | Cloud services, CI/CD topologies, incident root-cause chains, and SLOs |
|
| 66 |
+
| [`VaidhyaMegha/edge-ai-kg`](https://huggingface.co/datasets/VaidhyaMegha/edge-ai-kg) | Edge accelerators, quantization profiles, latency-power envelopes, and operators |
|
| 67 |
+
| [`VaidhyaMegha/wildfire-evac-kg`](https://huggingface.co/datasets/VaidhyaMegha/wildfire-evac-kg) | Emergency road networks, hazard zones, shelter capacities, and evacuation routing |
|
| 68 |
+
| [`VaidhyaMegha/edtech-kg`](https://huggingface.co/datasets/VaidhyaMegha/edtech-kg) | Prerequisite concept graphs, curriculum standards, and mastery trajectories |
|
| 69 |
+
|
| 70 |
+
### Sports Analytics Knowledge Graphs
|
| 71 |
+
| Dataset | Domain Focus |
|
| 72 |
+
| :--- | :--- |
|
| 73 |
+
| [`VaidhyaMegha/cricket-kg`](https://huggingface.co/datasets/VaidhyaMegha/cricket-kg) | `21K+` matches, players, venues, ball-by-ball statistics (`1.4M+` edges) |
|
| 74 |
+
| [`VaidhyaMegha/football-kg`](https://huggingface.co/datasets/VaidhyaMegha/football-kg) | Clubs, competitions, player transfers, tactical roles, and match events |
|
| 75 |
+
|
| 76 |
+
---
|
| 77 |
+
|
| 78 |
+
## 3. The Shoonya Initiative ([`VaidhyaMegha/Shoonya`](https://huggingface.co/VaidhyaMegha/Shoonya))
|
| 79 |
+
|
| 80 |
+
*Shoonya* (Sanskrit: शून्य — *zero / the foundational origin*) is our from-scratch, open-architecture Language & Code Model research program. Rather than treating small language model training as an opaque loss curve, **Shoonya** pairs a clean PyTorch transformer backbone with a **zero-dependency 4-crate Rust systems & mechanistic telemetry engine (`shoonya_rs`)**:
|
| 81 |
+
|
| 82 |
+
- **Modernized Compact Backbone:** `Pre-RMSNorm`, `SwiGLU`, `Grouped-Query Attention (GQA)`, `Rotary Position Embeddings (RoPE)`, **Low-Rank Factorized Tied Token Embeddings** (cutting vocabulary parameter overhead by ~`74.9%` in sub-500M models), and hardware-fused `BF16` FlashAttention-2 (`SDPA`).
|
| 83 |
+
- **Live Rust Mechanistic & Spectral Telemetry (`shoonya-telemetry`):** Real-time Jacobi SVD tracking of **Heavy-Tailed Self-Regularization (`HTSR α ∈ [2.0, 4.0]`)**, **Von Neumann Effective Rank (`eRank`)**, per-head **Normalized Attention Entropy**, **Attention Sink Ratio**, **Jensen-Shannon Head Redundancy**, and an **EWMA + MAD Statistical Anomaly Circuit Breaker** with closed-loop spectral weight-decay control.
|
| 84 |
+
- **Sandboxed Execution-Guided Evaluation & Closed-Loop Curriculum (`shoonya-eval`):** Live subprocess execution measuring unbiased `pass@1`, `pass@k`, and `pass^k` (multi-trial determinism), an 8-category AST failure taxonomy, Expected Calibration Error (`ECE`), and dynamic mid-training domain re-packing (`shoonya-core`).
|
| 85 |
+
- **Staged Scaling Ladder:**
|
| 86 |
+
- **`v0.4.0` (`17.16M` params):** CPU-trainable algorithmic wind-tunnel baseline (`64.1 MB` checkpoint).
|
| 87 |
+
- **`Stage 1` — `Shoonya-Code-135M` (`142.46M` params):** `d=768, L=12, H=12, KV=4` — **`81.2%` Stage-A `pass@1`**, **`85.4%` `pass@2`**, **`78.1%` `pass^2`**, `HTSR α = 1.97` (`Ideal_HeavyTailed_SOTA`), `78.8%` effective rank, `0` syntax/indentation errors across sandboxed generations.
|
| 88 |
+
- **`Stage 2` — `Shoonya-Code-350M` (`~266M` factorized / `350M`-class backbone):** `d=1024, L=20, H=16, KV=4, r_embed=256` with extended sequence context.
|
| 89 |
+
- **`Stage 3` — `Shoonya-Code-1.3B` (Roadmap):** Multi-language Python + Rust + Cypher/GraphQL code & graph-query synthesis integrated directly with **Samyama Graph**.
|
| 90 |
+
|
| 91 |
+
---
|
| 92 |
+
|
| 93 |
+
## 4. Connecting Graph Intelligence & Small Language Models
|
| 94 |
+
|
| 95 |
+
Our overarching research thesis unites **Samyama Graph**, the **19 Open Knowledge Graphs**, and **Shoonya**:
|
| 96 |
+
1. **Small, Mechanistically Verified Models + Exact Graph Memory > Monolithic Hallucinating Models:** Compact, calibrated models (`135M–1.3B`) paired with deterministic `Cypher` / `MCP` retrieval over Samyama Knowledge Graphs deliver verifiable, auditable domain reasoning on edge and enterprise hardware.
|
| 97 |
+
2. **Graph-Aware Embeddings & Cypher Synthesis:** Future Shoonya stages target native Cypher query generation and topology-preserving graph-text embeddings trained across our 19 public Knowledge Graphs.
|