IsingBreaker / README.md
brscftc's picture
Update README.md
46e803e verified
---
license: other
license_name: brsx-open-license
license_link: https://brsxlabs.gt.tc/brsxlicense.html
pipeline_tag: zero-shot-classification
tags:
- Ising
- quantum
- phy
- physics
---
# IsingBreaker
## Overview
IsingBreaker is an experimental symbolic sequence classification model developed by BRSX-Labs.
The model analyzes sequences composed of four symbolic tokens:
```text
U D + -
```
and estimates the degree of structural order present within the sequence.
The goal is not language modeling, but pattern recognition, periodicity detection, and symbolic structure analysis.
---
## Classification Labels
### Absolute
Perfect repeating motifs and highly ordered structures.
Examples:
```text
UDUDUDUDUDUDUDUD...
UD+-UD+-UD+-UD+...
UU++DD--UU++DD--...
```
---
### Maybe
Mostly ordered structures containing small local perturbations.
Examples:
```text
UDUDUDUDUDDDUDUD...
UD+-UD++UD+-UD+-...
UU++DD--UU+DDD--...
```
---
### NoN
Chaotic or non-periodic structures.
Examples:
```text
U+D--DU+U-+D++UD...
+-U-++UU+D+-DDUU...
```
---
## Architecture
IsingBreaker uses a hybrid Mixture-of-Experts architecture composed of four independent expert branches:
### CNN Expert
Captures local motifs and short-range symbolic structures.
Specialized for:
* Local repetition
* Motif detection
* Symbol blocks
---
### GRU Expert
Captures sequential dependencies and order-sensitive patterns.
Specialized for:
* Temporal relationships
* Sequence continuity
* Ordered transitions
---
### Transformer Expert
Captures long-range interactions between distant symbols.
Specialized for:
* Global structure
* Long-distance dependencies
* Pattern consistency
---
### Mamba Expert
Provides efficient state-space sequence modeling.
Specialized for:
* Long-context symbolic reasoning
* Efficient memory retention
* Sequence compression
---
## Expert Fusion
Outputs from all four experts are combined through a learned gating mechanism.
The model dynamically allocates attention between experts depending on the structure of the input sequence.
Example expert activity:
```text
CNN 0.28
GRU 0.24
Transformer 0.22
Mamba 0.26
```
---
## Model Information
* Architecture: GenoLiteHybrid
* Parameters: ~88 Million
* Context Length: 64
* Vocabulary Size: 4
* Classes: 3
---
## Dataset
Training dataset:
* 1,500 Absolute samples
* 1,500 Maybe samples
* 1,500 NoN samples
Total:
```text
4,500 unique samples
```
All samples are unique and shuffled before training.
---
## Performance
Benchmark Accuracy:
```text
93%+
```
The model demonstrates reliable separation between:
* Fully ordered structures
* Partially corrupted structures
* Chaotic structures
while generalizing to unseen motif combinations.
---
## Example
Input:
```text
UDUD-UDUD+UDUD-UDUD+UDUD-UDUD+
```
Prediction:
```text
Absolute
```
Confidence:
```text
0.94+
```
---
## License
brsx-open-license
---
## Author
BRSX-Labs