-
Beyond LLM-as-a-Judge: Deterministic Metrics for Multilingual Generative Text Evaluation
Paper • 2604.05083 • Published -
QCRI/OmniScore-mxbai
Feature Extraction • 0.3B • Updated • 29 -
QCRI/OmniScore-deberta-v3
Feature Extraction • 0.2B • Updated • 188 • 1 -
QCRI/OmniScore-Data
Viewer • Updated • 581k • 251 • 1
AI & ML interests
None defined yet.
Recent Activity
Papers
Fanar-Sadiq: A Multi-Agent Architecture for Grounded Islamic QA
From RAG to Agentic RAG for Faithful Islamic Question Answering
Organization Card
Qatar Computing Research Institute (QCRI), part of Hamad Bin Khalifa University, conducts applied and foundational research in computing to address national, regional, and global challenges. Our work spans Arabic language technologies, multilingual NLP, speech and conversational AI, social computing, data analytics, cybersecurity, and responsible AI. Here we host QCRI’s open models, datasets, and benchmarks to support reproducible research and advance AI technologies for Arabic, low-resource languages, and culturally grounded applications.
-
Beyond LLM-as-a-Judge: Deterministic Metrics for Multilingual Generative Text Evaluation
Paper • 2604.05083 • Published -
QCRI/OmniScore-mxbai
Feature Extraction • 0.3B • Updated • 29 -
QCRI/OmniScore-deberta-v3
Feature Extraction • 0.2B • Updated • 188 • 1 -
QCRI/OmniScore-Data
Viewer • Updated • 581k • 251 • 1
A comprehensive Arabic-centric multimodal generative AI stack with specialized models for text, image and poetry generation and image understanding
models 38
QCRI/MemeLens-VLM
Image-Text-to-Text • 9B • Updated • 573 • 2
QCRI/LlamaLens
Text Generation • 8B • Updated • 41 • 2
QCRI/OmniScore-mxbai
Feature Extraction • 0.3B • Updated • 29
QCRI/distillHarness
Updated
QCRI/OmniScore-deberta-v3
Feature Extraction • 0.2B • Updated • 188 • 1
QCRI/Fanar-2-Oryx-IG
Text-to-Image • Updated • 7 • 2
QCRI/Fanar-2-Oryx-IVU
Image-Text-to-Text • 8B • Updated • 37 • 4
QCRI/Fanar-2-Diwan
Text Generation • Updated • 1.34k • 2
QCRI/Fanar-2-27B-Instruct
Text Generation • 27B • Updated • 3.96k • 11
QCRI/MemeIntel
Image-Text-to-Text • 11B • Updated • 7
datasets 57
QCRI/ImageEval2026-Task1
Viewer • Updated • 16k • 1.07k • 1
QCRI/BloomBench
Viewer • Updated • 8.01k • 55
QCRI/ArGuard-Task1
Viewer • Updated • 4.5k • 41
QCRI/DisasterVQA
Viewer • Updated • 4.41k • 1.09k • 1
QCRI/WASIL
Viewer • Updated • 2.4k • 2.3k • 1
QCRI/Fanar-Sadiq-Bench
Viewer • Updated • 2.42k • 56
QCRI/MemeLens
Updated • 3.48k • 4
QCRI/ArabicCulturalQA
Viewer • Updated • 53.8k • 231 • 1
QCRI/OASIS
Viewer • Updated • 140 • 714
QCRI/IslamicFaithQA
Viewer • Updated • 7.62k • 54 • 3