LiteSearch：用於LLM的高效樹搜索

LiteSearch: Efficacious Tree Search for LLM

June 29, 2024

作者: Ante Wang, Linfeng Song, Ye Tian, Baolin Peng, Dian Yu, Haitao Mi, Jinsong Su, Dong Yu

cs.AI

摘要

最近的研究表明，樹搜索算法（例如蒙特卡羅樹搜索）可以顯著提升在複雜數學推理任務上的LLM性能。然而，由於浪費性的搜索策略，它們通常需要超過貪婪解碼的10倍以上的計算資源，這使得它們難以應用於實際應用中。本研究引入了一種新穎的引導樹搜索算法，具有動態節點選擇和節點級探索預算（最大子節點數）計算，以應對這個問題。通過考慮搜索進展朝著最終答案（歷史）以及來自價值網絡（未來）的引導，在沒有任何逐步註釋的情況下訓練，我們的算法在分配的計算預算範圍內迭代地選擇最有前途的樹節點，然後對其進行擴展。在GSM8K和TabMWP數據集上進行的實驗表明，我們的方法不僅提供了有競爭力的性能，而且與基準方法相比，計算成本顯著降低。

English

Recent research suggests that tree search algorithms (e.g. Monte Carlo Tree Search) can dramatically boost LLM performance on complex mathematical reasoning tasks. However, they often require more than 10 times the computational resources of greedy decoding due to wasteful search strategies, making them difficult to be deployed in practical applications. This study introduces a novel guided tree search algorithm with dynamic node selection and node-level exploration budget (maximum number of children) calculation to tackle this issue. By considering the search progress towards the final answer (history) and the guidance from a value network (future) trained without any step-wise annotations, our algorithm iteratively selects the most promising tree node before expanding it within the boundaries of the allocated computational budget. Experiments conducted on the GSM8K and TabMWP datasets demonstrate that our approach not only offers competitive performance but also enjoys significantly lower computational costs compared to baseline methods.