This paper introduces an LLM-assisted active learning method for OWL ontologies, where subsumption queries are reformulated into verbalized counter-concepts for LLMs. LLMs provide real-world examples to approximate these counter-concepts, ensuring that only Type II errors occur, which merely delay the construction process without introducing inconsistencies.
The content introduces the challenges of optimizing sustainable aquaculture using AI, specifically citing data scarcity, privacy concerns, and the simulation-to-reality gap in computer vision applications. It describes the author's journey to formulate a Privacy-Preserving Active Learning approach with inverse simulation verification to address these practical issues.
This content explains how AI can automate literature reviews by using active learning to screen overwhelming piles of PDFs. It details how AI learns from user decisions and prioritizes records, employing uncertainty sampling to efficiently identify relevant studies.
The author recounts their struggle to develop a precision oncology model for rare pediatric sarcoma, facing extreme data sparsity (47 samples) and strict HIPAA/GDPR constraints that prevented data sharing across institutions. This personal journey underscores the critical need for privacy-preserving active learning to address these challenges in real-world clinical workflows.
This article describes a researcher's frustration with extreme data sparsity in circular manufacturing supply chains for rare-earth magnets. The research was sparked by a dilemma between collecting more data or forcing sharing, leading to an epiphany about active learning for rare-event detection and privacy preservation.
This research explores combining privacy-preserving machine learning, specifically differential privacy and active learning, for the maintenance of bio-inspired soft robotics. The work addresses the challenge of retraining predictive maintenance models without exposing proprietary data during critical recovery windows.
LEAP is a closed-loop framework combining a domain-specialized large language model (LLM) with active learning for iterative additive prioritization in perovskite solar cells. It extracts knowledge from literature and represents molecules for Bayesian optimization, outperforming general-purpose models and validated experimentally.