arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2310.01446cs.CLcs.AI
收藏

大型语言模型推理中用于动态策略选择的Adaptive-Solver框架

Adaptive-Solver Framework for Dynamic Strategy Selection in Large Language Model Reasoning

School of Computer Science and Engineering(计算机科学与工程学院);School of Software Engineering(软件学院);Sun Yat-sen University(中山大学)

Jianpeng Zhou, Wanjun Zhong, Yanlin Wang, Jiahai Wang

AI总结:

提出Adaptive-Solver框架,通过初始评估和自适应模块动态调整LLM推理策略,在保持性能的同时降低API成本高达85%,或在同等成本下提升4.5%准确率。

摘要:

大型语言模型(LLMs)在处理推理任务时展现出了令人印象深刻的能力。然而,与人类能够本能地根据任务复杂性调整问题解决策略不同,大多数基于LLM的方法采用一刀切的方式。这些方法无论问题复杂度如何,都采用一致的模型、样本量、提示方法和问题分解程度。这些方法的不灵活性会带来不必要的计算开销或次优性能。为了解决这一局限性,我们引入了Adaptive-Solver(AS)框架,该框架动态调整解决策略以适应各种问题,从而实现测试时计算资源的灵活分配。该框架由两个主要模块运作。初始评估模块使用答案一致性评估当前解决方案的可靠性。如果解决方案被认为不可靠,则后续的自适应模块就会发挥作用。在此模块内,各种类型的自适应策略被协同采用。通过这种动态和多方面的自适应,我们的框架可以帮助减少计算消耗并提高性能。来自复杂推理基准的实验结果表明,我们的方法可以显著降低API成本(高达85%),同时保持原有性能。或者,在相同成本下,与基线相比,它实现了高达4.5%的更高准确率。代码和数据集可在https://github.com/john1226966735/Adaptive-Solver获取。

英文摘要:

Large Language Models (LLMs) demonstrate impressive ability in handling reasoning tasks. However, unlike humans who can instinctively adapt their problem-solving strategies to the complexity of task, most LLM-based methods adopt a one-size-fits-all approach. These methods employ consistent models, sample sizes, prompting methods and levels of problem decomposition, regardless of the problem complexity. The inflexibility of these methods can bring unnecessary computational overhead or sub-optimal performance. To address this limitation, we introduce an Adaptive-Solver (AS) framework tha dynamically adapts solving strategies to suit various problems, enabling the flexible allocation of test-time computational resources. The framework functions with two primary modules. The initial evaluation module assesses the reliability of the current solution using answer consistency. If the solution is deemed unreliable, the subsequent adaptation module comes into play. Within this module, various types of adaptation strategies are employed collaboratively. Through such dynamic and multi-faceted adaptations, our framework can help reduce computational consumption and improve performance. Experimental results from complex reasoning benchmarks reveal that our method can significantly reduce API costs (up to 85%) while maintaining original performance. Alternatively, it achieves up to 4.5% higher accuracy compared to the baselines at the same cost. The code and dataset are available at https://github.com/john1226966735/Adaptive-Solver.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑