Publications

EvoLMM: Self-Evolving Large Multimodal Models with Continuous Rewards

CoVR-R: Reason-Aware Composed Video Retrieval

Thinking Beyond Labels: Vocabulary-Free Fine-Grained Recognition using Reasoning-Augmented LMMs

DuwatBench: Bridging Language and Visual Heritage through an Arabic Calligraphy Benchmark for Multimodal Understanding

Beyond Simple Edits: Composed Video Retrieval with Dense Modifications

LlamaV-o1: Rethinking Step-by-Step Visual Reasoning in LLMs

TimeTravel: A Benchmark for Historical & Cultural Artifact Understanding

MobiLLaMA: Towards Accurate & Lightweight Fully Transparent GPT

All Languages Matter: Evaluating LMMs on 100 Culturally Diverse Languages

Composed Video Retrieval via Enriched Context and Discriminative Embeddings

CAMEL-Bench: A Comprehensive Arabic LMM Benchmark

Fann or Flop: A Benchmark for Arabic Poetry Understanding

XrayGPT: Chest Radiograph Summarization using Medical VLMs

3D Mitochondria Instance Segmentation with Spatio-Temporal Transformers

Fast Video Instance Segmentation via Recurrent Encoder Transformers

Video Instance Segmentation via Multi-Scale Spatio-Temporal Attention

Video Instance Segmentation in an Open-World

Image and video super resolution using recurrent generative adversarial network

Motion saliency based generative adversarial network for underwater moving object segmentation

Mobile-O: Unified Multimodal Understanding & Generation on Mobile

LLM Post-Training: A Deep Dive into Reasoning Large Language Models

Ain: The arabic inclusive large multimodal model

Dynamic pre-training: Towards efficient and scalable all-in-one image restoration

System and Method for Video Instance Segmentation via Multi-Scale Spatio-Temporal Transformers