On Conditional and Compositional Language Model Differentiable Prompting

被引:0
|
作者
Pilault, Jonathan [1 ]
Liu, Can [2 ]
Bansal, Mohit [3 ]
Dreyer, Markus [2 ]
机构
[1] Polytech Montreal, Mila Quebec AI Inst, Montreal, PQ, Canada
[2] Amazon Alexa, Seattle, WA USA
[3] Univ North Carolina Chapel Hill, Chapel Hill, NC USA
来源
PROCEEDINGS OF THE THIRTY-SECOND INTERNATIONAL JOINT CONFERENCE ON ARTIFICIAL INTELLIGENCE, IJCAI 2023 | 2023年
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
Prompts have been shown to be an effective method to adapt a frozen Pretrained Language Model (PLM) to perform well on downstream tasks. Prompts can be represented by a human-engineered word sequence or by a learned continuous embedding. In this work, we investigate conditional and compositional differentiable prompting. We propose a new model, Prompt Production System (PROPS), which learns to transform task instructions or input metadata, into continuous prompts that elicit task-specific outputs from the PLM. Our model uses a modular network structure based on our neural formulation of Production Systems, which allows the model to learn discrete rules - neural functions that learn to specialize in transforming particular prompt input patterns, making it suitable for compositional transfer learning and few-shot learning. We present extensive empirical and theoretical analysis and show that PROPS consistently surpasses other PLM adaptation techniques, and often improves upon fully fine-tuned models, on compositional generalization tasks, controllable summarization and multilingual translation, while needing fewer trainable parameters.
引用
收藏
页码:4136 / 4144
页数:9
相关论文
共 50 条
  • [31] Benchmarking Large Language Model Capabilities for Conditional Generation
    Maynez, Joshua
    Agrawal, Priyanka
    Gehrmann, Sebastian
    PROCEEDINGS OF THE 61ST ANNUAL MEETING OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS (ACL 2023): LONG PAPERS, VOL 1, 2023, : 9194 - 9213
  • [32] Automatic item generation in various STEM subjects using large language model prompting
    Park, Joonhyeong (joonhyeong.park@nie.edu.sg), 2025, 8
  • [33] Compositional Operational Semantics of a UML-Kernel-Model Language
    Fecher, Harald
    Kyas, Marcel
    de Roever, Willem-Paul
    de Boer, Frank S.
    ELECTRONIC NOTES IN THEORETICAL COMPUTER SCIENCE, 2006, 156 (01) : 79 - 96
  • [34] Emotion Recognition in Conversation with Multi-step Prompting Using Large Language Model
    Hama, Kenta
    Otsuka, Atsushi
    Ishii, Ryo
    SOCIAL COMPUTING AND SOCIAL MEDIA, PT I, SCSM 2024, 2024, 14703 : 338 - 346
  • [35] Chinese Metaphor Recognition Using a Multi-stage Prompting Large Language Model
    Wang, Jie
    Wang, Jin
    Zhang, Xuejie
    NATURAL LANGUAGE PROCESSING AND CHINESE COMPUTING, PT V, NLPCC 2024, 2025, 15363 : 234 - 246
  • [36] Prompting disentangled embeddings for knowledge graph completion with pre-trained language model
    Geng, Yuxia
    Chen, Jiaoyan
    Zeng, Yuhang
    Chen, Zhuo
    Zhang, Wen
    Pan, Jeff Z.
    Wang, Yuxiang
    Xu, Xiaoliang
    EXPERT SYSTEMS WITH APPLICATIONS, 2025, 268
  • [37] Enhancing large language model capabilities for rumor detection with Knowledge-Powered Prompting
    Yan, Yeqing
    Zheng, Peng
    Wang, Yongjun
    ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE, 2024, 133
  • [38] CLIP-SP: Vision-language model with adaptive prompting for scene parsing
    Li, Jiaao
    Huang, Yixiang
    Wu, Ming
    Zhang, Bin
    Ji, Xu
    Zhang, Chuang
    COMPUTATIONAL VISUAL MEDIA, 2024, 10 (04) : 741 - 752
  • [39] Demo: Accelerating Patient Screening for Clinical Trials using Large Language Model Prompting
    Gopeekrishnan, Anand
    Arif, Shibbir Ahmed
    Liu, Hao
    2024 IEEE/ACM CONFERENCE ON CONNECTED HEALTH: APPLICATIONS, SYSTEMS AND ENGINEERING TECHNOLOGIES, CHASE 2024, 2024, : 214 - 215
  • [40] Differentiable Particle Filters through Conditional Normalizing Flow
    Chen, Xiongjie
    Wen, Hao
    Li, Yunpeng
    2021 IEEE 24TH INTERNATIONAL CONFERENCE ON INFORMATION FUSION (FUSION), 2021, : 168 - 173