Adapting multi-objective pure-exploration bandits enables efficient Pareto prompt set recovery and best feasible prompt identification for LLMs, with linear-case guarantees and empirical gains over baselines.
For the multi-objective bandit setting, we have the prompt or arm setXwith|X |=K, and the expected performance or reward vector µ(x), x∈ X
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
-
Efficient Multi-objective Prompt Optimization via Pure-exploration Bandits
Adapting multi-objective pure-exploration bandits enables efficient Pareto prompt set recovery and best feasible prompt identification for LLMs, with linear-case guarantees and empirical gains over baselines.