Researcher

Mohaddeseh Tabrizian

Mohaddeseh Tabrizian is a graduate student in computer science and a research assistant at DFKI, the German Research Center for AI, with a focus on computer vision and natural language processing. After studying computer science at the University of Tabriz from 2016 to 2021, they began master's studies at RPTU Kaiserslautern-Landau in 2023. In May 2025, they co-authored an arXiv paper showing that story-based priming changes how LLM agents negotiate in a public goods game; the work was presented at NETGCOOP 2025 and published by Springer in 2026. They taught at RPTU as a lecturer in December 2025 and as a workshop instructor in March 2026. In DFKI's Data Science and its Applications group, they work as a research assistant and own an LLM-based document-understanding tool used by an industry partner.

Now at
DFKI

Timeline

  1. 2016–2021

    BSc at the University of Tabriz

    They study computer science at the University of Tabriz from 2016 to 2021 and earn a BSc.

  2. Since 2023

    Master's studies at RPTU

    They begin master's studies in computer science at RPTU Kaiserslautern-Landau in 2023. Their thesis, Optimized DNN Inference via LLM-Guided Evolutionary Search, uses an LLM as a mutator for hardware-aware neural network optimization.

  3. May 2025

    Narrative priming for LLM agents

    They co-author an arXiv paper, submitted in May 2025, on whether shared stories can nudge LLM agents toward collaboration in a finitely repeated public goods game. The experiments show that story-based priming significantly affects negotiation strategies and success rates, and that common stories improve collaboration.

  4. 2025–2026

    Paper at NETGCOOP 2025

    The work is presented at NETGCOOP 2025 and at the 1st International Electronic Conference on Games. Springer publishes it in 2026 as "The Power of Stories: Narrative Priming in Networked Multi-Agent LLM Interactions" in Lecture Notes in Computer Science, volume 16173.

  5. December 2025–March 2026

    Teaching generative AI at RPTU

    They lecture on RAG and knowledge graphs in RPTU's Engineering with Generative AI course in December 2025 and instruct the Coding with Generative AI workshop in March 2026. They also serve as a teaching assistant for the course, which has more than 200 students.

  6. As of October 2026

    Research assistant at DFKI

    They work as a research assistant at DFKI, the German Research Center for AI, in the Data Science and its Applications group, with a focus on computer vision and natural language processing. They build a key information extraction pipeline that lifts F1 by about 15 percentage points over the prior baseline, and they own an LLM-based document-understanding tool used by an industry partner.

Career

  1. NowDFKIResearch assistant
  2. 2025–2026RPTU University Kaiserslautern-LandauLecturer, workshop instructor, and teaching assistant

Education

  1. From 2023RPTU University Kaiserslautern-LandauMaster’s studies in computer science
  2. 2021University of TabrizBSc

Sources and credits

Photo: We have no published photo of Mohaddeseh Tabrizian yet. Send us a free-licensed photo.

  1. Mohaddeseh Tabrizian — Data Science and its Applications dsa.dfki.de. Checked 21 October 2026.
  2. Mohaddeseh-Tabrizian (Mohaddeseh Tabrizian) · GitHub github.com. Checked 21 October 2026.
  3. [2505.03961] The Power of Stories: Narrative Priming Shapes How LLM Agents Collaborate and Compete arxiv.org. Checked 21 October 2026.
  4. Islam Mesabah · PhD student, RPTU & DFKI islammesabah.github.io. Checked 21 October 2026.
  5. Team — Data Science and its Applications dsa.dfki.de. Checked 21 October 2026.