Sepideh Sadeghi
Papers
2
Total Citations
13
H-Index
2
About
Sepideh Sadeghi’s research lies at the intersection of cognitive science, artificial intelligence, and language acquisition, with a focus on how agents—both human and artificial—learn word meanings from ambiguous, real-world contexts. Her major contributions center on developing computational models of cross-situational and crossmodal word learning, addressing a fundamental challenge: how to incrementally map words to referents when multiple unknown words and objects co-occur. In her most-cited work, “An embodied incremental Bayesian model of cross-situational word learning” (2017, 9 citations), Sadeghi introduced a probabilistic framework that mirrors infant learning, enabling AI systems to resolve referential uncertainty over time. She extended this in “Models of Cross-Situational and Crossmodal Word Learning in Task-Oriented Scenarios” (2020, 4 citations), proposing dual models that co-learn object-word mappings and speaker intention from limited data. Her work bridges cognitive modeling and practical AI, offering scalable solutions for robots and virtual agents to acquire language in noisy, task-driven environments. Sadeghi’s research is notable for its theoretical rigor and applied relevance, making her a rising voice in grounded language learning.
Research Focus
Key Achievements
Top Papers
- 1
- 2