Model card: unify INT8 size and parity, drop baseline wording

README.md

@@ -56,11 +56,11 @@ pipeline_tag: token-classification
 
 A 38.7M-parameter byte-level Conv-Transformer that restores diacritics, tones, and vocalization marks across 37 languages without subword tokenizers or dictionary lookups: Akan (`ak`), Arabic (`ar`), Azerbaijani (`az`), Catalan (`ca`), Czech (`cs`), Welsh (`cy`), Ewe (`ee`), Spanish (`es`), Pulaar (`ff`), French (`fr`), Irish (`ga`), Guaraní (`gn`), Hausa (`ha`), Hebrew (`he`), Croatian (`hr`), Haitian Creole (`ht`), Hungarian (`hu`), Igbo (`ig`), Kurdish (`ku`), Lingala (`ln`), Lithuanian (`lt`), Latvian (`lv`), Māori (`mi`), Polish (`pl`), Portuguese (`pt`), Quechua (`qu`), Romanian (`ro`), Slovak (`sk`), Slovenian (`sl`), Samoan (`sm`), Serbian (`sr`), Turkmen (`tk`), Turkish (`tr`), Uzbek (`uz`), Vietnamese (`vi`), Wolof (`wo`), and Yorùbá (`yo`).
 
-On the official academic held-out **Yorùbá YAD test set** (3,330 sentences, 142k characters), it achieves a **15.88% Diacritic Error Rate (DER)**, **19.38% Word Error Rate (WER)**, and **5.58% Character Error Rate (CER)** with **0.0139% text corruption** (zero invented or dropped words), improving over classical baselines by 53.22 points.
+On the official academic held-out **Yorùbá YAD test set** (3,330 sentences, 142k characters), it achieves a **15.88% Diacritic Error Rate (DER)**, **19.38% Word Error Rate (WER)**, and **5.58% Character Error Rate (CER)** with **0.0139% text corruption** (zero invented or dropped words), against 69.10% DER for the unmarked input.
 
 Across the 37-language joint evaluation suite, it achieves **93.69% macro marked-position accuracy** with a composite score of **0.8419**. Thirteen languages reach 100% Exact Match and 0.00% CER on benchmark probes.
 
-The model is exported into native on-device formats: a **41.75 MB INT8 ONNX graph** for CPU and WebAssembly, and a compiled **Core ML package** for the Apple Neural Engine. Quantized INT8 matches full-precision PyTorch with **99.40% character parity** across 828 evaluation characters.
+The model is exported into native on-device formats: a **41.77 MB INT8 ONNX graph** for CPU and WebAssembly, and a compiled **Core ML package** for the Apple Neural Engine. Quantized INT8 matches full-precision PyTorch with **99.64% character parity** (822 of 825 characters over the 74 evaluation probes).
 
 ---
 
@@ -136,10 +136,10 @@ Held-out evaluation report across all 37 languages:
 
 | Format | Precision | File Size | Recommended Target | Latency (CPU / Apple NE) |
 | :--- | :--- | ---: | :--- | :---: |
-| `mark_int8.onnx` | Dynamic INT8 | **41.75 MB** | Edge CPU, Mobile, Browser (WASM) | 71.49 ms (4-thread CPU) |
-| `mark_fp16.onnx` | Float16 | **80.01 MB** | Mobile GPUs, WebGPU | 25.10 ms (GPU) |
+| `mark_int8.onnx` | Dynamic INT8 | **41.77 MB** | Edge CPU, Mobile, Browser (WASM) | 71.49 ms (4-thread CPU) |
+| `mark_fp16.onnx` | Float16 | **80.06 MB** | Mobile GPUs, WebGPU | 25.10 ms (GPU) |
 | `mark.mlpackage` | 8-bit Core ML | **42.10 MB** | Apple Neural Engine (iOS, macOS) | 12.40 ms (ANE) |
-| `mark_fp32.onnx` | Float32 | **158.81 MB** | Reference server baseline | 135.15 ms (1-thread CPU) |
+| `mark_fp32.onnx` | Float32 | **158.84 MB** | Reference server baseline | 135.15 ms (1-thread CPU) |
 
 ---
 
@@ -265,9 +265,9 @@ const spanish = await mark.restore("El nino comio jamon en la manana.", "es", {
 
 | File | Format | Size | Description |
 | :--- | :--- | ---: | :--- |
-| `mark_int8.onnx` | ONNX (INT8) | 41.75 MB | Dynamic INT8 quantized graph for CPU, Mobile, and WebAssembly |
-| `mark_fp16.onnx` | ONNX (FP16) | 80.01 MB | Half-precision graph for GPUs and Neural Engines |
-| `mark_fp32.onnx` | ONNX (FP32) | 158.81 MB | Full-precision reference model |
+| `mark_int8.onnx` | ONNX (INT8) | 41.77 MB | Dynamic INT8 quantized graph for CPU, Mobile, and WebAssembly |
+| `mark_fp16.onnx` | ONNX (FP16) | 80.06 MB | Half-precision graph for GPUs and Neural Engines |
+| `mark_fp32.onnx` | ONNX (FP32) | 158.84 MB | Full-precision reference model |
 | `mark.mlpackage.zip` | Core ML | 36.16 MB | Compiled Core ML package for Apple Neural Engine |
 | `config.json` | JSON | 1 KB | Model architectural hyperparameters |
 | `tags.json` | JSON | 40 KB | 1,073 tag operation mappings |