Search papers, labs, and topics across Lattice.
University of Maryland
5
0
9
Cognitive episodes in LRM reasoning traces reveal that item difficulty is shaped more by problem-solving dynamics than by item text alone.
AgentX can autonomously iterate on recommendation algorithms, outpacing human-driven processes and fundamentally changing how we approach system development.
LLMs can identify some discrimination signals in assessment items, but their predictions fall significantly short of human benchmarks.
Adaptive weighting in model merging can drastically improve multilingual reasoning performance, outperforming traditional methods across 21 languages.
A VLM can autonomously evolve its questioning capabilities, producing harder and more diverse questions that enhance its overall performance without needing external data.
