arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2602.00885cs.LG
收藏

可靠性感知的行列式点过程用于大型语言模型中鲁棒的信息数据选择

Reliability-Aware Determinantal Point Processes for Robust Informative Data Selection in Large Language Models

Clemson University, Clemson, SC, USA(克林姆恩大学);School of Computing, Clemson University, Clemson, SC, USA(计算学院)

Ahmad Sarlak, Abolfazl Razi

AI总结:

本文提出ProbDPP,一种可靠性感知的行列式点过程,用于在不确定环境下实现鲁棒的信息数据选择,以提高大型语言模型的部署效率。

摘要:

信息数据选择是大型语言模型(LLMs)的关键需求,以最小化微调、网络蒸馏和令牌修剪所需的数据量,从而实现快速高效的部署,尤其是在计算和通信限制下。传统子集选择方法,包括基于行列式点过程(DPP)的方法,专注于最大化多样性,但假设所选数据批次始终无错误地可用。这种假设禁止了在部分存储故障、不完美通信和随机访问故障情况下的使用。此外,我们显示在这些条件下原始公式会崩溃。为解决这一差距,我们引入了ProbDPP,一种新的可靠性感知的k-DPP实现,通过将目标函数重新表述为包含正则化项的公式来考虑概率数据访问,该公式保持良好定义,并分解为几何多样性项和不可靠成本项。所得到的目标函数在不确定性下促进了多样化的数据批次选择。此外,我们将这种可靠性感知的多样性最大化框架为组合半带问题,并提出了一种UCB类型的算法,以高效地在线学习未知的可靠性。理论分析为所提出的方法提供了后悔界,确保了性能保证。

英文摘要:

Informative data selection is a key requirement for large language models (LLMs) to minimize the amount of data required for fine-tuning, network distillation, and token pruning, enabling fast and efficient deployment, especially under computational and communication constraints. Traditional subset selection methods, including those based on Determinantal Point Processes (DPP), focus on maximizing diversity but assume that selected data batches are always available error-free. This presumption prohibits their use under partial storage outage, imperfect communication, and stochastic access failures. Furthermore, we show that the original formulation collapses under such conditions. To address this gap, we introduce ProbDPP, a novel reliability-aware implementation of k-DPP that accounts for probabilistic data access by recasting the objective function with a regularization term that remains well-posed and decomposes into a geometric diversity term and unreliability cost. The resulting objective facilitates robust selection of diverse data batches under uncertainty. Furthermore, we frame this reliability-aware diversity maximization as a combinatorial semi-bandit problem and propose a UCB-style algorithm to efficiently learn the unknown reliability online. Theoretical analysis provides regret bounds for the proposed approach, ensuring performance guarantees.

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。

↑