Search papers, labs, and topics across Lattice.
University of Southern California
3
0
6
14
Task order can dramatically skew the performance of self-improving agents, revealing hidden dependencies that could undermine their reliability in real-world applications.
Memory-augmented speculation boosts LLM prediction accuracy by up to 39% without incurring any additional execution time.
On-policy distillation makes language models more accurate, but also dangerously overconfident, revealing a fundamental tension between capability and calibration.
