Update multilingual v2 model card

README.md

@@ -6,6 +6,10 @@ datasets:
   - ai4privacy/pii-masking-200k
   - ai4privacy/pii-masking-400k
   - ai4privacy/open-pii-masking-500k-ai4privacy
+  - ai4privacy/pii-masking-openpii-1m
+  - nvidia/Nemotron-PII
+  - gretelai/gretel-pii-masking-en-v1
+  - piimb/privy
 pipeline_tag: token-classification
 tags:
   - token-classification
@@ -41,13 +45,14 @@ Fine-tuned [`openai/privacy-filter`](https://huggingface.co/openai/privacy-filte
 for **fine-grained PII extraction** across **54 categories** in **16 languages**.
 This v2 checkpoint is the more performant successor to
 `OpenMed/privacy-filter-multilingual`, with stronger multilingual PII masking
-behavior while keeping the same 16-language, fine-grained OpenMed label space.
+behavior while keeping the same 16-language, fine-grained OpenMed label space
+and runtime interface.
 
 - **Base model**: [`openai/privacy-filter`](https://huggingface.co/openai/privacy-filter) — 1.4B-parameter MoE (50M active per token), BIOES token-classification head
 - **Task**: Token classification for PII detection (BIOES scheme)
 - **Languages (16)**: Arabic, Bengali, Chinese, Dutch, English, French, German, Hindi, Italian, Japanese, Korean, Portuguese, Spanish, Telugu, Turkish, Vietnamese
-- **Training data**: Multilingual mix from [AI4Privacy](https://huggingface.co/ai4privacy) — `pii-masking-200k`, `pii-masking-400k`, and `open-pii-masking-500k-ai4privacy`, language-balanced
-- **Recipe**: `opf train` (OpenAI's official fine-tuning CLI) — full fine-tune, AdamW, balanced language sampling, 5 epochs, bf16
+- **Training data**: The original language-balanced multilingual OpenMed/AI4Privacy mix, followed by a v2 source-balanced privacy-masking adaptation mix from AI4Privacy OpenPII, Nemotron, Gretel, and Privy-style PII data
+- **Recipe**: `opf train` (OpenAI's official fine-tuning CLI) — full fine-tune, AdamW, balanced language and source sampling, bf16
 - **Labels**: 54 PII categories → 217 BIOES classes (1 `O` + 54 × B/I/E/S)
 
 The base model ships with 8 coarse PII categories and English-only training. This
@@ -55,10 +60,12 @@ model trades that for a **6.75× more granular vocabulary** spanning identity,
 contact, address, financial, vehicle, digital, and crypto labels — all evaluated
 across 16 languages.
 
-> **Family at a glance.** Same architecture, three runtimes:
-> - **PyTorch (this repo)** — CPU + CUDA, anywhere transformers runs.
-> - **MLX BF16** — [`OpenMed/privacy-filter-multilingual-mlx`](https://huggingface.co/OpenMed/privacy-filter-multilingual-mlx) — Apple Silicon, full precision.
-> - **MLX 8-bit** — [`OpenMed/privacy-filter-multilingual-mlx-8bit`](https://huggingface.co/OpenMed/privacy-filter-multilingual-mlx-8bit) — Apple Silicon, smaller + faster.
+> **Runtime note.** This v2 upload is the PyTorch checkpoint for CPU/CUDA
+> inference anywhere `transformers` runs. The existing MLX repositories
+> [`OpenMed/privacy-filter-multilingual-mlx`](https://huggingface.co/OpenMed/privacy-filter-multilingual-mlx)
+> and [`OpenMed/privacy-filter-multilingual-mlx-8bit`](https://huggingface.co/OpenMed/privacy-filter-multilingual-mlx-8bit)
+> are first-generation multilingual siblings; use this repo when you
+> specifically want v2 behavior until v2 MLX conversions are published.
 
 ## Quick start
 
@@ -66,8 +73,7 @@ across 16 languages.
 
 OpenMed gives you `extract_pii()` / `deidentify()` with built-in BIOES Viterbi
 decoding, span refinement, and a Faker-backed obfuscation engine. Same call
-on every host — Apple Silicon picks up MLX automatically; everywhere else uses
-this PyTorch checkpoint.
+on every host that supports this PyTorch checkpoint.
 
 ```bash
 pip install -U "openmed[hf]"
@@ -82,30 +88,30 @@ text = (
 )
 
 # Extract grouped entity spans
-result = extract_pii(text, model_name="OpenMed/privacy-filter-multilingual")
+result = extract_pii(text, model_name="OpenMed/privacy-filter-multilingual-v2")
 for ent in result.entities:
     print(f"{ent.label:30s} {ent.text!r}  conf={ent.confidence:.2f}")
 
 # De-identify with any of the supported methods
-masked   = deidentify(text, method="mask",   model_name="OpenMed/privacy-filter-multilingual")
-removed  = deidentify(text, method="remove", model_name="OpenMed/privacy-filter-multilingual")
-hashed   = deidentify(text, method="hash",   model_name="OpenMed/privacy-filter-multilingual")
+masked   = deidentify(text, method="mask",   model_name="OpenMed/privacy-filter-multilingual-v2")
+removed  = deidentify(text, method="remove", model_name="OpenMed/privacy-filter-multilingual-v2")
+hashed   = deidentify(text, method="hash",   model_name="OpenMed/privacy-filter-multilingual-v2")
 
 # Faker-backed locale-aware obfuscation, deterministic with consistent=True+seed
 fake = deidentify(
     text,
     method="replace",
-    model_name="OpenMed/privacy-filter-multilingual",
+    model_name="OpenMed/privacy-filter-multilingual-v2",
     consistent=True,
     seed=42,
 )
 print(fake.deidentified_text)
 ```
 
-`OpenMed/privacy-filter-multilingual-mlx*` model names also work in the same
-`extract_pii()` / `deidentify()` calls — on a non-Apple-Silicon host they
-automatically fall back to **this PyTorch checkpoint** with a one-time warning.
-So you can ship MLX names in code and still run on Linux/Windows.
+Use `OpenMed/privacy-filter-multilingual-v2` in `extract_pii()` /
+`deidentify()` when you want this v2 checkpoint. The first-generation
+`OpenMed/privacy-filter-multilingual-mlx*` model names remain available for
+Apple Silicon workflows, but they are separate artifacts.
 
 The OpenMed wrapper passes `trust_remote_code=True` for you, runs the model's
 own BIOES Viterbi decoder, and skips OpenMed's regex smart-merging (the model
@@ -180,12 +186,12 @@ If you use this model, please cite **this model**, the organization behind it
 (**OpenMed**), and the upstream base model + datasets:
 
 ```bibtex
-@misc{openmed_privacy_filter_multilingual_2026,
+@misc{openmed_privacy_filter_multilingual_v2_2026,
   author       = {OpenMed},
-  title        = {{OpenMed/privacy-filter-multilingual}: multilingual fine-grained PII extraction across 16 languages and 54 categories},
+  title        = {{OpenMed/privacy-filter-multilingual-v2}: multilingual fine-grained PII extraction across 16 languages and 54 categories},
   year         = {2026},
   publisher    = {Hugging Face},
-  howpublished = {\url{https://huggingface.co/OpenMed/privacy-filter-multilingual}}
+  howpublished = {\url{https://huggingface.co/OpenMed/privacy-filter-multilingual-v2}}
 }
 
 @misc{openmed_2026,