Ilya Safro

The paper introduces a method to explain biomedical hypotheses using graph-based context retrieval and LLMs, with feedback loops to refine explanations.

This paper introduces AGATHA, an AI tool for drug repurposing in dementia by analyzing biomedical literature and identifying gene-disease associations using statistical methods.

The paper introduces an interpretable visualization approach for scientific hypotheses using semantic concept interconnectivity, network-based methods, and topic modeling, validated with a pipeline evaluated by experts.

The paper introduces two hypothesis generation systems for COVID-19 research using graph mining and transformer models, achieving high-quality predictions and discovering new research findings.

This study identifies DDX3 inhibition as a potential therapy for HAND by blocking Tat and cocaine-induced microglial activation and neurotoxicity in rat and mouse models.

This paper validates and ranks biomedical hypothesis generation systems using automated metrics and a validation challenge, demonstrating the utility of their approach with the MOLIERE system.