Search papers, labs, and topics across Lattice.
2
0
4
1
LLMs exhibit a pervasive optimism bias when evaluating research proposals, frequently rating methodologically unsound ideas as promising, suggesting they're not ready to replace human reviewers.
Evaluative language in LLM-generated resume summaries can introduce significant biases that traditional fairness audits might miss, destabilizing hiring processes.
