Learning to Forget: Emotional Salience as a Compression Mechanism for Long-Term AI Memory
Open Access
- 8 August 2026
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 13259-13264
Abstract
No abstract availableFunding Information
- Ministry of Science and ICT, South Korea (26-InnoCORE-01)
- Ministry of Science and ICT, South Korea (RS-2025-02219277)
- Ministry of Science and ICT, South Korea (N10250156)
- Ministry of Science and ICT, South Korea (02-26-01-0057)
This publication has 11 references indexed in Scilit:
- Unable to forget: Proactive lnterference Reveals Working Memory Limits in LLMs Beyond Context LengthOSF Preprints, 2025
- Lost in the Middle: How Language Models Use Long ContextsTransactions of the Association for Computational Linguistics, 2024
- Learning, Forgetting, Remembering: Insights From Tracking LLM Memorization During Training2024
- The amygdala mediates the facilitating influence of emotions on memory through multiple interacting mechanismsNeurobiology of Stress, 2023
- ATOMIC: An Atlas of Machine Commonsense for If-Then ReasoningProceedings of the AAAI Conference on Artificial Intelligence, 2019
- The Biology of Forgetting—A PerspectiveNeuron, 2017
- The Persistence and Transience of MemoryNeuron, 2017
- Making lasting memories: Remembering the significantProceedings of the National Academy of Sciences, 2013
- Decay happens: the role of active forgetting in memoryTrends in Cognitive Sciences, 2013
- THE AMYGDALA MODULATES THE CONSOLIDATION OF MEMORIES OF EMOTIONALLY AROUSING EXPERIENCESAnnual Review of Neuroscience, 2004