Publications
Four articles across LLM-driven portfolio selection and data-centric deep learning for medical image segmentation.
Hierarchical Cluster Tournaments select stocks qualitatively by clustering assets by correlation and using LLM-run tournaments based on long-term fundamentals, producing transparent and diversified portfolios without optimization.
Read paperLLMs act as listwise tournament judges to rank large stock universes, combining active learning and probabilistic aggregation to produce stable, out-of-sample rankings.
Read paperAlphaLoss uses LLMs to autonomously discover robust, interpretable portfolio optimization losses through evolutionary refinement.
Read paperData-centric strategies enrich training data for mammography segmentation through annotation-guided intensity manipulation and style transfer, improving generalization across imaging equipment.
Read paperIdeas
Long-form notes on agent systems, security, and model infrastructure.
Setting Up a Secure Server for OpenClaw