Hamza Mohammed

efficient vision-language models

memory-efficient transformers

long-context processing

multi-agent systems

automated software testing

test case generation

software quality assurance

multi-agent communication

agent evaluation protocols

context retrieval

knowledge integration in AI systems

multi-modal retrieval

graph-based retrieval systems

Large Language Models

Preference Optimization

Fine-Tuning

sparse attention mechanisms

dynamic sparse attention

adaptive KV-cache compression

edge device deployment

on-device inference

hardware-aware model optimization

efficient transformer architectures

resource-efficient agent-based approaches

constrained VRAM environments

efficient inference

Image and Video Segmentation

ViTs

Model Distillation and Pruning

Object Tracking

4D compact representation