Robust Prompt Optimization for Large Language Models Against Distribution Shifts

被引:0
|
作者
Li, Moxin [1 ]
Wang, Wenjie [1 ]
Feng, Fuli [2 ,3 ]
Cao, Yixin [4 ]
Zhang, Jizhi [2 ]
Chua, Tat-Seng [1 ]
机构
[1] Natl Univ Singapore, Singapore, Singapore
[2] Univ Sci & Technol China, Hefei, Peoples R China
[3] Inst Dataspace, Hefei, Anhui, Peoples R China
[4] Singapore Management Univ, Singapore, Singapore
关键词
D O I
暂无
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Large Language Model (LLM) has demonstrated significant ability in various Natural Language Processing tasks. However, their effectiveness is highly dependent on the phrasing of the task prompt, leading to research on automatic prompt optimization using labeled task data. We reveal that these prompt optimization techniques are vulnerable to distribution shifts such as subpopulation shifts, which are common for LLMs in real-world scenarios such as customer reviews analysis. In this light, we propose a new problem of robust prompt optimization for LLMs against distribution shifts, which requires the prompt optimized over the labeled source group can simultaneously generalize to an unlabeled target group. To solve this problem, we propose Generalized Prompt Optimization framework, which incorporates the unlabeled data from the target group into prompt optimization. Extensive experimental results demonstrate the effectiveness of the proposed framework with significant performance improvement on the target group and comparable performance on the source group.
引用
收藏
页码:1539 / 1554
页数:16
相关论文
共 50 条
  • [1] Prompt Optimization in Large Language Models
    Sabbatella, Antonio
    Ponti, Andrea
    Giordani, Ilaria
    Candelieri, Antonio
    Archetti, Francesco
    MATHEMATICS, 2024, 12 (06)
  • [2] WORDFLOW: Social Prompt Engineering for Large Language Models
    Wang, Zijie J.
    Chakravarthy, Aishwarya
    Munechika, David
    Chau, Duen Horng
    PROCEEDINGS OF THE 62ND ANNUAL MEETING OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS, VOL 3: SYSTEM DEMONSTRATIONS, 2024, : 42 - 50
  • [3] Prompt Engineering: Unleashing the Power of Large Language Models to Defend Against Social Engineering Attacks
    Nezer, Ahmed I.
    Nema, Bashar M.
    Salim, Wisam Makki
    Iraqi Journal for Computer Science and Mathematics, 2024, 5 (03): : 404 - 416
  • [4] PromSec: Prompt Optimization for Secure Generation of Functional Source Code with Large Language Models (LLMs)
    Nazzal, Mahmoud
    Khalil, Issa
    Khreishah, Abdallah
    Phan, NhatHai
    CCS 2024 - Proceedings of the 2024 ACM SIGSAC Conference on Computer and Communications Security, : 2266 - 2279
  • [5] Black-Box Prompt Optimization: Aligning Large Language Models without Model Training
    Cheng, Jiale
    Liu, Xiao
    Zheng, Kehan
    Ke, Pei
    Wang, Hongning
    Dong, Yuxiao
    Tang, Jie
    Huang, Minlie
    PROCEEDINGS OF THE 62ND ANNUAL MEETING OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS, VOL 1: LONG PAPERS, 2024, : 3201 - 3219
  • [6] Federated Learning of Large Language Models with Parameter-Efficient Prompt Tuning and Adaptive Optimization
    Che, Tianshi
    Liu, Ji
    Zhou, Yang
    Ren, Jiaxiang
    Zhou, Jiwen
    Sheng, Victor S.
    Dai, Huaiyu
    Dou, Dejing
    2023 CONFERENCE ON EMPIRICAL METHODS IN NATURAL LANGUAGE PROCESSING, EMNLP 2023, 2023, : 7871 - 7888
  • [7] To prompt or not to prompt: Navigating the use of Large Language Models for integrating and modeling heterogeneous data
    Remadi, Adel
    El Hage, Karim
    Hobeika, Yasmina
    Bugiotti, Francesca
    DATA & KNOWLEDGE ENGINEERING, 2024, 152
  • [8] <monospace>BadCodePrompt</monospace>: backdoor attacks against prompt engineering of large language models for code generation
    Qu, Yubin
    Huang, Song
    Li, Yanzhou
    Bai, Tongtong
    Chen, Xiang
    Wang, Xingya
    Li, Long
    Yao, Yongming
    AUTOMATED SOFTWARE ENGINEERING, 2025, 32 (01)
  • [9] PromptMaker: Prompt-based Prototyping with Large Language Models
    Jiang, Ellen
    Olson, Kristen
    Toh, Edwin
    Molina, Alejandra
    Donsbach, Aaron
    Terry, Michael
    Cai, Carrie J.
    EXTENDED ABSTRACTS OF THE 2022 CHI CONFERENCE ON HUMAN FACTORS IN COMPUTING SYSTEMS, CHI 2022, 2022,
  • [10] Balancing Privacy and Robustness in Prompt Learning for Large Language Models
    Shi, Chiyu
    Su, Junyu
    Chu, Chiawei
    Wang, Baoping
    Feng, Duanyang
    MATHEMATICS, 2024, 12 (21)