Visiting Research Intern at TIB Hannover
Working with Prof. Sören Auer on provenance and attribution methods for the Open Research Knowledge Graph.
My doctoral research develops IKnowFlow, a framework for making the flow of knowledge interpretable and auditable in AI systems through interpretable retrieval, explainable evidence selection, and grounded source attribution.
I am a Ph.D. student in Computer Science at the University of Maryland, Baltimore County (UMBC). My research focuses on making AI systems trustworthy for sensitive domains. I formalize IKnowFlow or Interpretable Knowledge Flow, as a framework for making knowledge flow auditable across stages: why a document was retrieved, why it was selected as evidence, and whether a generated claim can be traced back to that evidence. My work spans interpretable dense retrieval with IMRNNs, rationale-driven evidence selection with METEORA, attribution benchmarking with REASONS, and provenance-aware verification using structured scholarly knowledge graphs.
A linked timeline of current research updates, talks, and coverage.
Working with Prof. Sören Auer on provenance and attribution methods for the Open Research Knowledge Graph.
IKnowFlow was accepted to the IJCAI-ECAI 2026 Doctoral Consortium in Bremen, Germany, with travel grant support.
METEORA reframes re-ranking as rationale-driven evidence selection for safer and more interpretable RAG in sensitive domains.
Coverage of attribution-based reasoning results comparing OpenAI o1 and DeepSeek R1, including hallucination behavior on sentence-level citation tasks.
IMRNNs introduces embedding modulation for interpretable dense retrieval and exposes human-readable query-document matching signals.