LEVI shows that stronger evolutionary search architecture can match or beat frontier-LLM performance at 3-35x lower cost across systems and prompt-optimization benchmarks.
25- For larger sets, use greedy selection based on compression-like utility (sum of squares of frequencies)
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.NE 1years
2026 1verdicts
UNVERDICTED 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
LEVI: Stronger Search Architectures Can Substitute for Larger LLMs in Evolutionary Search
LEVI shows that stronger evolutionary search architecture can match or beat frontier-LLM performance at 3-35x lower cost across systems and prompt-optimization benchmarks.