ReVEL evolves more robust and diverse heuristics for combinatorial optimization by embedding LLMs as multi-turn reasoners that analyze grouped performance feedback within an evolutionary meta-controller.
In the context ofLLM-driven optimization, the variation operator is re-parameterized as a conditional generation task using a Large Language Model πθ
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
ReVEL: Multi-Turn Reflective LLM-Guided Heuristic Evolution via Structured Performance Feedback
ReVEL evolves more robust and diverse heuristics for combinatorial optimization by embedding LLMs as multi-turn reasoners that analyze grouped performance feedback within an evolutionary meta-controller.