Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
Data Science Institute, London School of Economics and Political Science, Department of Mathematics, London School of Economics and Political Science
2
0
4
Certifying missed relevant items in high-recall candidate generation requires sampling from the excluded pool, revealing a surprising minimax rate-optimal label complexity.
The worst-case displacement of model representations can be precisely quantified using a generalized eigenvalue approach, revealing critical vulnerabilities in sentiment classifiers.