RESEARCH

Research

I study how visual AI systems reason, generalise and report evidence they can be held to — and build the software that puts those methods in front of people.

• BSc in Computer Science and Engineering, East Delta University, Chattogram, Bangladesh
• CGPA 3.79 / 4.00 · Expected graduation February 2027
• Seeking a research-focused, preferably funded MSc starting Fall 2027, as the first step towards doctoral research.

Interests: Vision-language models · Computer vision · Multimodal learning · Visual grounding · Trustworthy AI · Robustness and generalisation · Explainable AI

Research projects

• InsightVLM — Naming the statistical finding a chart shows — and checking it against the plotted data. Manuscript in preparation.
• FoodGenome AI — A 97% food classifier that also says, per image, when not to trust it. Completed research project.
• RoboDesign Atlas — Domain-adapting CLIP to retrieve engineering figures, and auditing what the corpus contains. Completed research project.
• FairShift-PFN — Fairness side-effects of label-free test-time adaptation in tabular foundation models. Manuscript in preparation.
• CauseGround — Visual grounding and reliable multimodal reasoning. Ongoing research.

Research projects & manuscripts

There are no published, accepted, submitted or preprint papers yet.

• InsightVLM-20K: chart images paired with statistically verified findings — Manuscript in preparation
• What Can Be Certified Without Labels? Fairness Side-Effects of Test-Time Adaptation in Tabular Foundation Models — Manuscript in preparation