RPS applies reinforcement learning to adaptively select prompts that elicit concealed information in LLM dialogues and outperforms static baselines on a new legal-case benchmark called IELegal.
• Environment Interaction.Generate a batch of candidate lawyer prompts ( Z(s)) for the same conversation state s, and sequentially conduct multiple rounds of dialogue
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
-
RPS: Information Elicitation with Reinforcement Prompt Selection
RPS applies reinforcement learning to adaptively select prompts that elicit concealed information in LLM dialogues and outperforms static baselines on a new legal-case benchmark called IELegal.