BayanSimplify-v0.2
0
3 commits
1 repos using this model
1 linked in READMEs
updated Oct 5, 2026
Model 2 of the Bayan project: AraT5v2 trained on a 22,733-pair mix with strength tags.
Trained on SAMER, Baseet and DAASI pairs, each tagged with the style it teaches: [S0] SAMER (minimal word swaps),
[S1] light, [S2] medium, [S3] strong (Baseet levels 3, 2, 1), [SA] everyday and administrative (DAASI). The tag
sets the strength at run time; with no tag the model is most careful with meaning.
Input: بسّط: (with the shadda), then an optional tag such as [S2] , then the text.
BayanBench v2.0, test, core items (meaning kept = Gemma 4 31B P(same) ≥ 0.5 and every number kept):
| Setting | Meaning kept | Longest clause, words cut |
|---|---|---|
[S2], float | 59.5% | 6.3 |
| no tag, float | 68.3% | 4.5 |
| no tag, int8 in the app | 78.6% | 1.9 |
It was the app's large model until BayanSimplify-v0.3 replaced it, and stays available as "Earlier large model".
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
repo = "Congi-libya/BayanSimplify-v0.2"
tok = AutoTokenizer.from_pretrained(repo)
model = AutoModelForSeq2SeqLM.from_pretrained(repo)
text = "وقد أدى التوسع العمراني السريع الذي شهدته المدينة خلال العقدين الماضيين إلى ازدحام مروري خانق في ساعات الذروة."
x = tok("بسّط: " + text, return_tensors="pt", max_length=256, truncation=True)
y = model.generate(**x, num_beams=1, no_repeat_ngram_size=3, max_length=256)
print(tok.decode(y[0], skip_special_tokens=True))
| Model | What it is | Base model | Training data | In the Bayan app |
|---|---|---|---|---|
| BayanSimplify-v0.1 | Model 1 | AraT5v2-base-1024 | SAMER, level 5 → 3 | not shipped |
| BayanSimplify-v0.2 | Model 2 | AraT5v2-base-1024 | 22,733-pair mix (SAMER, Baseet, DAASI) with strength tags | "Earlier large model" |
| BayanSimplify-v0.2-Fast | Model 2, compact | AraBART | the same mix | "Fast model", the default download |
| BayanSimplify-v0.3 | Model 3, our best | AraT5v2-base-1024 | Bayan corpus v1, 14,975 rows | "Large model", four beams |
| BayanSimplify-ONNX | The int8 bundles the app downloads | v0.2-Fast, v0.2, v0.3 |
CC BY-NC 4.0 (non-commercial), in line with the training data's terms. SAMER is used under the CAMeL Lab's permission to fine-tune and share weights for non-commercial use; its text is not redistributed here.
Bayan (بيان) simplifies Arabic text for readers with dyslexia, entirely on an Android phone: select text in any app, choose «تبسيط» (Simplify), and a simpler version appears over the page. Built by Team Cogni for the Samsung Innovation Campus AI capstone, 2026. Every number on this card comes from the team's final report and BayanBench v2.0, the benchmark built for the task (meaning scored by Gemma 4 31B, checked against human raters, AUC 0.85).
BayanSimplify-v0.2
0
3 commits
1 repos using this model
1 linked in READMEs
updated Oct 5, 2026
Model 2 of the Bayan project: AraT5v2 trained on a 22,733-pair mix with strength tags.
Trained on SAMER, Baseet and DAASI pairs, each tagged with the style it teaches: [S0] SAMER (minimal word swaps),
[S1] light, [S2] medium, [S3] strong (Baseet levels 3, 2, 1), [SA] everyday and administrative (DAASI). The tag
sets the strength at run time; with no tag the model is most careful with meaning.
Input: بسّط: (with the shadda), then an optional tag such as [S2] , then the text.
BayanBench v2.0, test, core items (meaning kept = Gemma 4 31B P(same) ≥ 0.5 and every number kept):
| Setting | Meaning kept | Longest clause, words cut |
|---|---|---|
[S2], float | 59.5% | 6.3 |
| no tag, float | 68.3% | 4.5 |
| no tag, int8 in the app | 78.6% | 1.9 |
It was the app's large model until BayanSimplify-v0.3 replaced it, and stays available as "Earlier large model".
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
repo = "Congi-libya/BayanSimplify-v0.2"
tok = AutoTokenizer.from_pretrained(repo)
model = AutoModelForSeq2SeqLM.from_pretrained(repo)
text = "وقد أدى التوسع العمراني السريع الذي شهدته المدينة خلال العقدين الماضيين إلى ازدحام مروري خانق في ساعات الذروة."
x = tok("بسّط: " + text, return_tensors="pt", max_length=256, truncation=True)
y = model.generate(**x, num_beams=1, no_repeat_ngram_size=3, max_length=256)
print(tok.decode(y[0], skip_special_tokens=True))
| Model | What it is | Base model | Training data | In the Bayan app |
|---|---|---|---|---|
| BayanSimplify-v0.1 | Model 1 | AraT5v2-base-1024 | SAMER, level 5 → 3 | not shipped |
| BayanSimplify-v0.2 | Model 2 | AraT5v2-base-1024 | 22,733-pair mix (SAMER, Baseet, DAASI) with strength tags | "Earlier large model" |
| BayanSimplify-v0.2-Fast | Model 2, compact | AraBART | the same mix | "Fast model", the default download |
| BayanSimplify-v0.3 | Model 3, our best | AraT5v2-base-1024 | Bayan corpus v1, 14,975 rows | "Large model", four beams |
| BayanSimplify-ONNX | The int8 bundles the app downloads | v0.2-Fast, v0.2, v0.3 |
CC BY-NC 4.0 (non-commercial), in line with the training data's terms. SAMER is used under the CAMeL Lab's permission to fine-tune and share weights for non-commercial use; its text is not redistributed here.
Bayan (بيان) simplifies Arabic text for readers with dyslexia, entirely on an Android phone: select text in any app, choose «تبسيط» (Simplify), and a simpler version appears over the page. Built by Team Cogni for the Samsung Innovation Campus AI capstone, 2026. Every number on this card comes from the team's final report and BayanBench v2.0, the benchmark built for the task (meaning scored by Gemma 4 31B, checked against human raters, AUC 0.85).