Update model card

README.md

@@ -19,41 +19,13 @@ Every LoRA ships as `<Name>_KREA2.safetensors` alongside `<Name>_KREA2.md` insid
 
 ### Repository restructure
 - All characters moved under a single **`Characters/`** directory (one folder per character containing the `.safetensors`, the persona `.md`, and the training captions).
-- All **`Images_Prompt_Examples`** folders and their contents were **removed**.
 - A new **`Pruned_Comparison`** folder was added with one preview image per character (see below).
 
 ### Dynamic re-rank + block pruning
-Every LoRA was re-ranked with **dynamic `sv_fro` SVD** (retention `0.998`, max rank `16`) and block-pruned by dropping the **8 lowest-impact transformer blocks** (blocks `1-8`, ~3% of delta energy). All `txtfusion` refiner/layerwise keys are preserved. Weights are stored as `lora_up` / `lora_down` + `alpha` in fp16, loadable directly in ComfyUI. Each file carries `crt_*` provenance metadata.
-
-### Trigger word corrections
-27 persona `.md` files had trigger words that did not match the tokens actually used in their training captions. These were corrected to the exact training token, e.g.:
-
-| Character | Was | Now |
-|---|---|---|
-| Adriana Lima | `Adriana Lima` | `AdrianaLima` |
-| Anne Hathaway | `Anne Hathaway` | `AnneHathaway` |
-| Ariana Grande | `Ariana Grande` | `ArianaGrande` |
-| Billie Eilish | `Billie_ilish` | `BillieEilish` |
-| Doja Cat | `Dojcat` | `DojaCat` |
-| Elydia | `unknown` | `Helydia` |
-| Eva Mendes | `unknown` | `EvaMendes` |
-| Ho-yeon | `ho-yeon` | `JungHoYeon` |
-| Jessica Alba | `Jessica_lba` | `JessicaAlba` |
-| Kylie Jenner | `Kylie_enner` | `KylieJenner` |
-| Lea Elui | `Léa-elui` | `LeaElui` |
-| Nicki Minaj | `NickiM` | `NickiMinaj` |
-| Sab Zada | `SZA` | `SabZada` |
-| Shakira | `Sh_akira` | `Shakira` |
-| Shay | `Vanessa Lesnicki` | `Shay` |
-| Tokyo | `TKY` | `Tokyo` |
-| Zoe Kravitz | `Zkravit` | `ZoeKravitz` |
-
-(plus Alexandria Ocasio-Cortez, Ana De Armas, Anya Taylor-Joy, Bella Poarch, Camille Cerf, Dua Lipa, Emilia Clarke, Hannah Kleit, Inde Navarrette, Kaia Gerber)
+Every LoRA was re-ranked with **dynamic `sv_fro` SVD** (retention `0.998`, max rank `16`) and block-pruned by dropping the **8 lowest-impact transformer blocks** (blocks `1-8`, ~3% of delta energy).
 
 ### Retrained character
-- **Jung Ho-yeon** (`Ho-yeon`) was **retrained** on a new dataset, then re-ranked and pruned like the others. Its trigger word is now **`JungHoYeon`**.
-
-> Note: `Autumn Falls` and `Hailey Rose` have no matching training dataset, so their trigger word remains `unknown`.
+- **Jung Ho-yeon** (`Ho-yeon`) was **retrained** on a new dataset, then re-ranked and pruned like the others.
 
 ## Fully automated pipeline