AIMS

Artificial Intelligence & Multimodal Systems

Lab @ USM · CSCE

A research group at the University of Southern Mississippi building AI systems that learn from images, video, text, speech, and sensor data, and stay reliable under uncertainty.

Projects

Same Ranking, Different Winner

The scoring target a memory benchmark picks can flip which LLM wins, even when the overall ranking looks stable.

#ConversationalMemory#Evaluation

Fixed RAG Compression

A fixed compression layer in a RAG pipeline collapses the reader-scaling trend that the uncompressed setup shows.

#RAG#Evaluation

Outcome Monitors

Recovery affordances that let an agent notice and repair tool calls that failed silently.

#AIAgents#TrustworthyAI

Thin-Object Segmentation

A cross-domain survey of power-line, crack, and retinal-vessel segmentation, from classical pipelines to promptable foundation models.

#ComputerVision#FoundationModels

SEAM

Measuring how much typed user speech gets absorbed into edited artifacts at unmarked within-turn seams.

#HumanCenteredAI#LLMEditing

People

Faculty

  • Rabab Abdelfattah Rabab Abdelfattah Principal Investigator

PhD Students

  • Olanrewaju Muili Olanrewaju Muili Trustworthy AI
  • Siyan Luo PhD student

Undergraduates

  • Sugam Panthi Sugam Panthi LLM evaluation validity
  • Muhaiminul Yeamin Muhaiminul Yeamin Research assistant
  • Sakshyam Sigdel Sakshyam Sigdel Research assistant
  • Nhoojah Maharjan Nhoojah Maharjan Research assistant
From Classical Pipelines to Promptable Foundation Models: A Cross-Domain Survey of Thin-Object Segmentation for Power Lines, Cracks, and Retinal Vessels A. Hossain, N. Maharjan, M. Hasan, R. Abdelfattah, M. Ezz-Eldin, X. Wang, M. M. Fouda, K. Abdelfattah · IEEE Internet of Things Journal · 2026 Same Ranking, Different Winner: How Scoring Targets Shape LLM Memory Benchmarks S. Panthi, R. Abdelfattah · Findings of EMNLP 2026 · 2026 Outcome Monitors: Recovery Affordances for Silent Tool Failures S. Panthi, R. Abdelfattah · arXiv preprint · 2026