Search papers, labs, and topics across Lattice.
3
0
7
Counterfactual recourse in education becomes truly actionable when recommendations are not just model-valid but also semantically feasible and machine-checkable.
Forget brittle, opaque tagging systems: this LLM pipeline aligns learning resources to competencies with high accuracy *and* provides human-auditable evidence for each tag.
Naive early prediction models can be dangerously overconfident: temporal leakage from assessment data can dramatically inflate reported performance.