TypoFormer (Stage 2 INT8)

High-Precision Character-Aware Hybrid Transformer for Edge Autocomplete & Non-Autoregressive Grammar Correction
_{Created & Authored by Md. Fahim Sarker Mridul}

Architecture | Highlights | Comparison | Benchmarks | Files | Quickstart | Windows Desktop | Citation

---

Overview

TypoFormer is a purpose-engineered, character-aware neural backbone designed from the ground up to solve two conflicting natural language problems inside a single unified 94M parameter trunk:

1. Causal Next-Word Autocomplete: Low-latency autoregressive token prediction under strict sub-15ms local keystroke budgets without cloud connectivity.
2. Non-Autoregressive Grammar & Typo Correction (GEC): High-precision edit-tag sequence labeling that repairs punctuation slips, confused vocabulary, subject-verb agreement, and severe typos without generative hallucination.

By fusing Byte-Level BPE subwords with Multi-Scale Character 1D-Convolutions through a learned gating highway, TypoFormer eliminates out-of-vocabulary token fractures and delivers enterprise-grade real-time writing assistance completely offline on standard desktop CPUs.

---

Model Architecture

Figure 1: TypoFormer Full Internal Neural Architecture: Fusing byte-level subwords with multi-scale character convolutions, Pre-RMSNorm grouped-query attention (GQA), rotary embeddings (RoPE), SwiGLU feed-forward networks, and multi-task dual heads for offline autocomplete and grammar error correction (Visualized in NVIDIA Emerald Green #76B900).
[View High-Resolution 300 DPI Architecture Diagram (PNG)]

1. Dual-Channel Gated Input Highway

Standard tokenizers fail when words are corrupted: mistyping "definitely" as "definatly" converts a single semantic token into multiple rare subword fragments. TypoFormer prevents this tokenization distortion through a hybrid input representation:

• Byte-Level BPE: Words are split into standard BPE tokens from a compact 16,000-vocabulary dictionary ($x_{\text{tok}} \to \mathbf{E}{\text{tok}} \in \mathbb{R}^{d{\text{model}}}$).
• Multi-Scale Character-CNN: Concurrently, the exact UTF-8 character bytes of each token ($x_{\text{char}}$) are fed into parallel 1D convolutional banks with kernel sizes $k \in {2, 3, 5, 8}$: $${\mathbf{C}}_k=\mathrm{M}\mathrm{a}\mathrm{x}\mathrm{P}\mathrm{o}\mathrm{o}\mathrm{l}\left(\mathrm{G}\mathrm{E}\mathrm{L}\mathrm{U}({\mathrm{C}\mathrm{o}\mathrm{n}\mathrm{v}1\mathrm{D}}_k(\text{Bytes}))\right)\backslash mathbf\{C\}\_k\; =\; \backslash mathrm\{MaxPool\}\backslash left(\backslash mathrm\{GELU\}\backslash left(\backslash mathrm\{Conv1D\}\_k(\backslash text\{Bytes\})\backslash right)\backslash right)$$
• Learned Gating Highway: $${\mathbf{h}}_0={\mathbf{E}}_{\text{tok}}+\alpha \cdot {\mathbf{W}}_{\text{proj}}([{\mathbf{C}}_2\mathrm{\parallel }{\mathbf{C}}_3\mathrm{\parallel }{\mathbf{C}}_5\mathrm{\parallel }{\mathbf{C}}_8])\backslash mathbf\{h\}\_0\; =\; \backslash mathbf\{E\}\_\{\backslash text\{tok\}\}\; +\; \backslash alpha\; \backslash cdot\; \backslash mathbf\{W\}\_\{\backslash text\{proj\}\}\backslash left([\backslash mathbf\{C\}\_2\; \backslash ,\backslash |\backslash ,\; \backslash mathbf\{C\}\_3\; \backslash ,\backslash |\backslash ,\; \backslash mathbf\{C\}\_5\; \backslash ,\backslash |\backslash ,\; \backslash mathbf\{C\}\_8]\backslash right)$$ where $\alpha$ is a learnable scalar gate initialized to ensure robust character-level fallback when unknown or miscoded subwords appear.

2. Transformer Trunk (10 Blocks, $d=768$, $d_{\text{ffn}}=2048$)

• Attention Mechanism: Grouped-Query Attention (GQA) with 12 Query heads and 4 Key-Value heads (3:1 ratio), cutting KV-cache memory traffic by 66% during sequence generation.
• Rotary Position Embeddings (RoPE): Frequency-based relative query-key rotation for superior relative token distance awareness across cursor movements without absolute position limits.
• Feed-Forward Backbone: SwiGLU non-linearity: $$\text{SwiGLU}(x)=W_2(\text{SiLU}(W_{1}x)\odot W_{3}x)\backslash text\{SwiGLU\}(x)\; =\; W\_2\backslash left(\backslash text\{SiLU\}(W\_1\; x)\; \backslash odot\; W\_3\; x\backslash right)$$
• Pre-Normalization: RMSNorm applied prior to each sub-layer for stable gradient propagation and reduced memory overhead.

3. Multi-Task Dual Output Heads

• Task 1: Causal LM Autocomplete Head: Linear projection ($d_{\text{model}} \to 16{,}000$, weight-tied with token embeddings) yielding next-token probability distribution $P(w_{t+1} \mid w_{\le t})$.
• Task 2: Keystroke GEC Edit Head: Linear tagging projection ($d_{\text{model}} \to 6{,}148$) mapping each token position to atomic edits:
  • KEEP (0): Token is grammatically and lexically correct.
  • DELETE (1): Redundant or erroneous token removal.
  • CASE_CAP (2) / CASE_LOWER (3): Casing adjustments.
  • REPLACE (4 .. 4099): Atomic swap with one of 4,096 high-frequency correction tokens.
  • APPEND (4100 .. 6147): Atomic right-insertion with one of 2,048 suffix/continuation tokens.

---

Key Highlights

• Ultra-Low Keystroke Latency: Mean end-to-end CPU inference of 9.97 ms (P95: 12.12 ms), fitting comfortably inside the 16.6 ms 60 FPS display frame deadline.
• Typo Immunity: Gated character-level convolutions preserve word roots even under severe phonetic and fat-finger keyboard slips.
• 100% Air-Gapped and Offline: Zero network calls, zero telemetry, zero risk of leaking confidential documents, code, or private emails.
• Guaranteed Content Fidelity: Non-autoregressive tagging guarantees zero generative hallucination and achieves 100.0% clean negative sentence preservation.
• Compact Edge Footprint: Symmetric INT8 quantization packs the complete 94M dual-task architecture into just 130.5 MB ONNX graphs.

---

Architectural Comparison: TypoFormer vs. BERT vs. Cloud LLMs

Architectural Dimension	Classical BERT-Base (110M)	Cloud Generative LLMs (7B - 70B)	TypoFormer (94M INT8)	TypoFormer Advantage
Generative Autocomplete	Incapable (Bidirectional MLM only)	Yes (High quality autoregression)	Native Causal Mode (Tied causal head)	Unified Trunk: Single model serves both autocomplete & GEC.
Typo & OOV Robustness	Brittle (teh to ['te', '##h'])	Fair (Relies on large BPE dictionary)	Gated Multi-Scale Char-CNN ($k=2,3,5,8$)	Typo Immunity: Never blinded by unknown character corruptions.
Position Encoding	Absolute Learned (Max 512 tokens)	RoPE / ALiBi	Rotary Position Embeddings (RoPE)	Sequence Extrapolation: Relative distance aware across cursor hops.
Attention Mechanism	Multi-Head (MHA) (12 Q, 12 K, 12 V)	GQA / MQA	Grouped Query Attention (GQA) (12 Q, 4 KV)	3x Memory Reduction: Cuts KV-cache bandwidth by 66%.
FFN Non-Linearity	GeLU + LayerNorm (Additive bias)	SwiGLU + RMSNorm	SwiGLU + Pre-RMSNorm	LLaMA/Mistral-Grade Efficiency: Higher expressivity per parameter.
GEC Formulation	Seq2Seq / MLM (Slow, iterative)	Generative Prompting (Prone to hallucinations)	Constrained Atomic Edit-Tagging (6,148 tags)	Zero Hallucination: 100% preservation of correct text.
End-to-End Latency	150ms - 400ms on CPU	300ms - 2,500ms via Cloud API	< 10ms on CPU (INT8 ONNX)	Real-Time Keystroke Budget: Executes within 16ms display frame.
Network & Privacy	Local or Cloud	Mandatory Cloud API (Data leaves host)	100% Offline Air-Gapped	Complete Privacy: Zero telemetry, password field suppression.

---

Benchmark and Evaluation Metrics

Live CPU benchmarks measured on standard Intel/AMD consumer hardware (scripts/evaluate_gec_live.py):

Metric	Measured Value	Benchmark Description
Mean CPU Latency	9.97 ms	Per-sentence end-to-end inference
P95 CPU Latency	12.12 ms	95th percentile latency (well below 16.6 ms display frame)
P99 CPU Latency	14.80 ms	99th percentile worst-case latency
GEC Stage 2 Validation Loss	0.2597	Down from 4.1067 (93.7% loss reduction across training)
LM Stage 2 Loss	1.2054	Fine-tuned on curated engineering and client communications
Clean Negative Preservation	100.0%	Zero false-positive edits on grammatically correct input text
GEC Model Footprint	130.5 MB	Symmetric INT8 ONNX graph (gec_stage2_int8.onnx)
LM Model Footprint	138.1 MB	Symmetric INT8 ONNX graph (lm_stage2_int8.onnx)
Active Working Set RAM	< 250 MB	Total memory footprint during active inference

---

Repository Contents

File	Size	Format	Description
gec_stage2_int8.onnx	130.5 MB	ONNX INT8	INT8 Quantized graph for grammar, punctuation, and typo correction
lm_stage2_int8.onnx	138.1 MB	ONNX INT8	INT8 Quantized graph for causal next-word autocomplete
edit_maps.json	228 KB	JSON	High-frequency confusion dictionary (4,096 replacements + 2,048 appends)
tokenizer.json	1.11 MB	JSON	Byte-Level BPE tokenizer config (16,000 vocabulary)
vocab.json / merges.txt	526 KB	Text/JSON	BPE vocabulary and merge table
config.json	520 B	JSON	Architecture parameters & model metadata
infer.py	6.5 KB	Python	Standalone, dependency-minimal Python inference engine
typoformer_architecture.png	235 KB	PNG (300 DPI)	Full neural architecture diagram (NVIDIA emerald theme)
typoformer_full_logo.png	188 KB	PNG	Official project branding logo
TypoFormer-win-x64.zip	221.0 MB	ZIP	Standalone Windows distribution package

---

Quickstart and Usage

1. Minimal Installation

pip install onnxruntime tokenizers numpy
git clone https://huggingface.co/fahiiim/TypoFormer
cd TypoFormer
python infer.py

2. Standalone Python Example

from infer import TypoFormer

# Initialize the standalone inference engine (loads ONNX models locally)
model = TypoFormer(".")

# 1. Autocomplete (Next-Word Generation)
prompt = "Could you please confirm if the proposed implementation timeline aligns with your"
suggestions = model.autocomplete(prompt, top_k=3)
print("Autocomplete suggestions:", suggestions)
# Output: [(' business', 0.9978), (' goals', 0.0002), (' company', 0.0001)]

# 2. Grammar, Punctuation & Typo Correction (GEC)
dirty_text = "We decoupled the event processing pipeline to achieve sub-"
suggestions = model.autocomplete(dirty_text, top_k=2)
print("Engineering completion:", suggestions)
# Output: [(' millisecond', 0.8841), (' second', 0.0712)]

# 3. Clean Text Preservation (Zero Hallucination)
clean_text = "The quick brown fox jumps over the lazy dog."
result = model.correct_grammar(clean_text)
assert result == clean_text  # 100% preservation

---

Windows Desktop Keystroke Companion (1-Line Install)

TypoFormer includes a native Windows x64 .NET 8 desktop daemon that runs silently in the system tray and renders a floating caret overlay across any active text editor (Word, Notepad, VS Code, Slack, Browser):

Run in PowerShell:

irm https://raw.githubusercontent.com/fahiiim/TyPoTaKer/main/install.ps1 | iex

• Offline & Private: Zero telemetry; no keystrokes ever leave your device.
• Password Aware: Automatically suppresses buffering on password and credential fields via Windows UI Automation.
• Instant Action: Press [Tab] to accept autocomplete suggestions or grammar corrections.

---

Citation and Author Attribution

@software{typoformer2026,
  author       = {Md. Fahim Sarker Mridul},
  title        = {TypoFormer: High-Precision Character-Aware Hybrid Transformer for Edge Autocomplete & Non-Autoregressive Grammar Correction},
  year         = {2026},
  publisher    = {Hugging Face},
  url          = {https://huggingface.co/fahiiim/TypoFormer},
  howpublished = {\url{https://github.com/fahiiim/TyPoTaKer}}
}

License

This model and its associated weights are distributed under the MIT License. Authored by Md. Fahim Sarker Mridul.