Research

Problems that matter here.

Urdu-language models, vision on sub-$50 hardware, evaluation standards for public deployment. Our labs work on what benchmarks ignore, and publish where it counts.

Labs
5
Publications
10
Active projects
4

Selected work

Recent publications.

  1. 2026
    ReportSafety, Alignment & Governance Lab

    Evaluating Language Models for Public-Sector Deployment: A Practical Guide

    Fatima Noor, Hamza Qureshi, Sana Malik · PIOAI Policy Report 2026-01

  2. 2026
    PaperLanguage & Reasoning Lab

    Measuring Reasoning Without Leaking the Test: Procedural Generation for LLM Evaluation

    Bilal Ahmed, Omar Farooq, Hamid Raza · ICLR 2026

  3. 2026
    PaperLanguage & Reasoning Lab

    UrduQA: A Contamination-Resistant Benchmark for Question Answering in Urdu

    Bilal Ahmed, Rafia Hussain, Ali Hassan, Sana Malik · ACL 2026 (Findings)

  4. 2025
    PaperPerception & Robotics Lab

    Field-Robust Cotton Leaf Disease Detection on Sub-$50 Hardware

    Zara Siddiqui, Imran Khalid, Noor Fatima Baig · CVPR 2025 Workshop on Computer Vision for Agriculture