CARE introduces an evidence-gated auditable controller for LLM-generated ranking policies in high-throughput experimentation, reporting performance gains on Minerva/Olympus (80.0 to 88.5) and ChemLex (83.9 to 92.1) benchmarks over the incumbent optimizer.
It must cover every candidate_df row with unique positive ranks and finite scores
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
CARE: Controlling LLM-Generated Policies through Auditable Review of Evidence in Scientific Experimentation
CARE introduces an evidence-gated auditable controller for LLM-generated ranking policies in high-throughput experimentation, reporting performance gains on Minerva/Olympus (80.0 to 88.5) and ChemLex (83.9 to 92.1) benchmarks over the incumbent optimizer.