Joana Plewnia

Papers

Total Citations

H-Index

About

Joana Plewnia is a researcher at the forefront of cognitive robotics, specializing in episodic memory systems that enable robots to learn from and communicate about their past experiences. Her work addresses a critical bottleneck in artificial intelligence: how robots can efficiently store, retrieve, and verbalize life-long experiences without being overwhelmed by data. Plewnia’s major contribution lies in developing hierarchical representations that allow robots to summarize and answer questions about their own history—a key step toward more natural human-robot interaction. In her 2024 paper "Forgetting in Robotic Episodic Long-Term Memory," she challenges the traditional approach of transferring all working memory data to long-term memory, proposing instead that selective forgetting can reduce data volumes while preserving essential information. Her complementary work on episodic memory verbalization (2024, 2025) demonstrates how robots can transform streams of episodic data into coherent narratives, bridging the gap between raw sensor data and human-understandable language. Though her citation counts are still growing (2 citations each for her most recent papers), Plewnia’s research is pioneering a new paradigm for lifelong robot learning—one where machines don’t just remember everything, but remember what matters.