Learning a Single Token to Replace Long System Prompts in LLMs
Open Access
- 6 November 2026
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 1048-1059
Abstract
No abstract availableFunding Information
- National Natural Science Foundation of China (62502404)
- Hong Kong Research Grants Council (Research Impact Fund) (R1015-23)
- Hong Kong Research Grants Council (Collaborative Research Fund) (C1043-24GF)
- Hong Kong Research Grants Council (RGC Research Fellow Scheme) (RFS2627-1S03)
- Hong Kong Research Grants Council (General Research Fund) (11218325)
- Hong Kong Research Grants Council (General Research Fund) (11212926)
- Institute of Digital Medicine of City University of Hong Kong (9229503)
- Huawei (Huawei Innovation Research Program) (-)
- Tencent (CCF-Tencent Open Fund) (-)
- Tencent (Tencent Rhino-Bird Focused Research Program) (-)
- Tencent (Tencent University Cooperation Project) (-)
- Kuaishou (CCF-Kuaishou Large Model Explorer Fund) (2025008)
- Kuaishou (Kuaishou University Cooperation Project) (-)
- Alibaba (CCF-Taobao & Tmall Group TECH Kangaroo Fund) (2026001)
- Bytedance (-)
This publication has 40 references indexed in Scilit:
- Contextual Attention Modulation: Towards Efficient Multi-Task Adaptation in Large Language Models2025
- Put Teacher in Student's Shoes: Cross-Distillation for Ultra-compact Model Compression Framework2025
- Large Multimodal Model Compression via Iterative Efficient Pruning and Distillation2024
- Efficient Memory Management for Large Language Model Serving with PagedAttention2023
- Prompt Distillation for Efficient LLM-based Recommendation2023
- PromptST: Prompt-Enhanced Spatio-Temporal Multi-Attribute Prediction2023
- Exploration and Regularization of the Latent Action Space in Recommendation2023
- Deep Reinforcement Learning for Information Retrieval: Fundamentals and Advances2020
- Deep reinforcement learning for page-wise recommendations2018
- Recommendations with Negative Feedback via Pairwise Deep Reinforcement Learning2018