A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.
Neural nets learn and forget tasks sequentially, showing promising scalability.
The paper explores updating multiple goals in reinforcement learning with neural networks.
Novel method diagnoses large language models' reasoning abilities.
SAKT improves knowledge tracing by focusing on relevant past activities.
Paper derives constraints for Bayesian Knowledge Tracing parameters.
The ERI is a new index for measuring exam readiness.
Bayesian model identifies skill difficulties and student subgroups in engineering education.
Analyzes how diffusion models learn, revealing a spectral bias in structure mastery.
A new system recommends educational content based on student and peer interests.