Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
# 🔮 DMind-3: The Age of Foresight
|
| 2 |
|
| 3 |
*From local logic to global foresight. In a world of isolated systems, the one who sees the whole board wins.*
|
|
@@ -38,6 +62,8 @@ This final piece of the trilogy moves beyond the tactical and into the strategic
|
|
| 38 |
|
| 39 |
DMind-3 introduces **Hierarchical Predictive Synthesis (HPS)**. While C³-SFT (used in `mini`) teaches the model to correct its own reasoning, HPS teaches it to synthesize multiple, conflicting, time-variant data streams into a coherent probabilistic forecast. It operates on a nested hierarchy of abstraction, from raw on-chain events to complex macroeconomic indicators.
|
| 40 |
|
|
|
|
|
|
|
| 41 |
**(Figure 1: The HPS training paradigm, showing multi-level data fusion and probabilistic future state generation)**
|
| 42 |
|
| 43 |
**Mathematical Formalization**
|
|
@@ -86,6 +112,8 @@ DMind-3 is designed to power the next generation of DeFi analytics, risk managem
|
|
| 86 |
|
| 87 |
The DMind-3 series is a vertically integrated stack designed for sovereign intelligence.
|
| 88 |
|
|
|
|
|
|
|
| 89 |
**(Figure 2: The full DMind-3 Cognitive Architecture, from on-device reflexes to cloud-native foresight)**
|
| 90 |
|
| 91 |
- **The Oracle (DMind-3)**: Runs in the cloud. Provides macro-strategic foresight, systemic risk analysis, and orchestrates the agent fleet.
|
|
@@ -143,4 +171,4 @@ print(response.choices[0].text)
|
|
| 143 |
- **Probabilistic Nature**: All forecasts are probabilistic and based on the data available up to the knowledge cutoff. The model cannot predict black swan events and is subject to the inherent unpredictability of markets.
|
| 144 |
- **Knowledge Cutoff**: The core model has a knowledge cutoff of June 2025. While it can process real-time data provided via the API, its foundational understanding is based on its training corpus.
|
| 145 |
|
| 146 |
-
---
|
|
|
|
| 1 |
+
---
|
| 2 |
+
language:
|
| 3 |
+
- en
|
| 4 |
+
- zh
|
| 5 |
+
license: apache-2.0
|
| 6 |
+
library_name: transformers
|
| 7 |
+
tags:
|
| 8 |
+
- web3
|
| 9 |
+
- finance
|
| 10 |
+
- defi
|
| 11 |
+
- chain-of-thought
|
| 12 |
+
- sft
|
| 13 |
+
- security-audit
|
| 14 |
+
- on-device-ai
|
| 15 |
+
metrics:
|
| 16 |
+
- accuracy
|
| 17 |
+
- ponzi-detection-rate
|
| 18 |
+
- code-security-score
|
| 19 |
+
pipeline_tag: text-generation
|
| 20 |
+
inference: false
|
| 21 |
+
base_model:
|
| 22 |
+
- openai/gpt-oss-20b
|
| 23 |
+
---
|
| 24 |
+
|
| 25 |
# 🔮 DMind-3: The Age of Foresight
|
| 26 |
|
| 27 |
*From local logic to global foresight. In a world of isolated systems, the one who sees the whole board wins.*
|
|
|
|
| 62 |
|
| 63 |
DMind-3 introduces **Hierarchical Predictive Synthesis (HPS)**. While C³-SFT (used in `mini`) teaches the model to correct its own reasoning, HPS teaches it to synthesize multiple, conflicting, time-variant data streams into a coherent probabilistic forecast. It operates on a nested hierarchy of abstraction, from raw on-chain events to complex macroeconomic indicators.
|
| 64 |
|
| 65 |
+

|
| 66 |
+
|
| 67 |
**(Figure 1: The HPS training paradigm, showing multi-level data fusion and probabilistic future state generation)**
|
| 68 |
|
| 69 |
**Mathematical Formalization**
|
|
|
|
| 112 |
|
| 113 |
The DMind-3 series is a vertically integrated stack designed for sovereign intelligence.
|
| 114 |
|
| 115 |
+

|
| 116 |
+
|
| 117 |
**(Figure 2: The full DMind-3 Cognitive Architecture, from on-device reflexes to cloud-native foresight)**
|
| 118 |
|
| 119 |
- **The Oracle (DMind-3)**: Runs in the cloud. Provides macro-strategic foresight, systemic risk analysis, and orchestrates the agent fleet.
|
|
|
|
| 171 |
- **Probabilistic Nature**: All forecasts are probabilistic and based on the data available up to the knowledge cutoff. The model cannot predict black swan events and is subject to the inherent unpredictability of markets.
|
| 172 |
- **Knowledge Cutoff**: The core model has a knowledge cutoff of June 2025. While it can process real-time data provided via the API, its foundational understanding is based on its training corpus.
|
| 173 |
|
| 174 |
+
---
|