Jasin Cekinmez
Hi, I’m Jasin, a senior at Princeton studying applied math and computer science. I build evaluations of multimodal and agentic AI systems, probing where they fail, why, and how their behavior diverges from humans. Grateful to be supported by a Laude Institute Slingshot. Check out my work below.
Along the way, I’ve interned on the Query Optimization team at Teradata and as an ML researcher at QCRI. When I’m not stress-testing models, I’m probably constructing crosswords 🧩 for The New York Times.
Research
SportD: Can VLMs Physically Strategize?
How does Adversarial Influence Scale in Multi-Agent Systems?
Quantifying Empirical Compute-Supervision Tradeoffs in RLVR
OccludeBench: Can VLMs Find What’s Missing?
Where Did the Ambiguity Go? Examining How Multimodal Models Interpret Polysemous Words
Query Timing Produces Opposite Positional Biases Between LLMs and Humans
ADAM: A Diverse Archive of Mankind for Evaluating and Enhancing LLMs in Biographical Reasoning
Guess the Unified Model: Domain and Linguistic Effects in Generated Images