A State-Update Prompting Strategy for Efficient and Robust Multi-turn Dialogue
This work addresses inefficiencies in long-horizon dialogues for LLM users, offering an incremental improvement through prompt engineering.
The paper tackles the problem of information forgetting and inefficiency in multi-turn dialogues with LLMs by proposing a training-free prompt engineering method, which improves core information filtering by 32.6% and reduces inference time by 73.1% on the HotpotQA dataset.
Large Language Models (LLMs) struggle with information forgetting and inefficiency in long-horizon, multi-turn dialogues. To address this, we propose a training-free prompt engineering method, the State-Update Multi-turn Dialogue Strategy. It utilizes "State Reconstruction" and "History Remind" mechanisms to effectively manage dialogue history. Our strategy shows strong performance across multiple multi-hop QA datasets. For instance, on the HotpotQA dataset, it improves the core information filtering score by 32.6%, leading to a 14.1% increase in the downstream QA score, while also reducing inference time by 73.1% and token consumption by 59.4%. Ablation studies confirm the pivotal roles of both components. Our work offers an effective solution for optimizing LLMs in long-range interactions, providing new insights for developing more robust Agents.