System Prompt Optimization with Meta-Learning

• URL: http://arxiv.org/abs/2505.09666v1
• Date: Wed, 14 May 2025 16:46:15 GMT
• Title: System Prompt Optimization with Meta-Learning
• Authors: Yumin Choi, Jinheon Baek, Sung Ju Hwang,
• Abstract summary: We introduce the novel problem of bilevel system prompt optimization, whose objective is to design system prompts that are robust to diverse user prompts.<n>We propose a meta-learning framework, which meta-learns the system prompt by optimizing it over various user prompts across multiple datasets.<n>We conduct experiments on 14 unseen datasets spanning 5 different domains, on which we show that our approach produces system prompts that generalize effectively to diverse user prompts.
• Score: 60.04718679054704
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large Language Models (LLMs) have shown remarkable capabilities, with optimizing their input prompts playing a pivotal role in maximizing their performance. However, while LLM prompts consist of both the task-agnostic system prompts and task-specific user prompts, existing work on prompt optimization has focused on user prompts specific to individual queries or tasks, and largely overlooked the system prompt that is, once optimized, applicable across different tasks and domains. Motivated by this, we introduce the novel problem of bilevel system prompt optimization, whose objective is to design system prompts that are robust to diverse user prompts and transferable to unseen tasks. To tackle this problem, we then propose a meta-learning framework, which meta-learns the system prompt by optimizing it over various user prompts across multiple datasets, while simultaneously updating the user prompts in an iterative manner to ensure synergy between them. We conduct experiments on 14 unseen datasets spanning 5 different domains, on which we show that our approach produces system prompts that generalize effectively to diverse user prompts. Also, our findings reveal that the optimized system prompt enables rapid adaptation even to unseen tasks, requiring fewer optimization steps for test-time user prompts while achieving improved performance.

Related papers

• P3: Prompts Promote Prompting [26.16464064171255]
  Large language model (LLM) applications often employ multi-component prompts, comprising both system and user prompts.<n>In this work, we introduce P3, a novel self-improvement framework that concurrently optimize both system and user prompts.<n>Extensive experiments on general tasks demonstrate that P3 achieves superior performance in the realm of automatic prompt optimization.
  arXiv  Detail & Related papers  (2025-07-21T14:37:46Z)
• Has My System Prompt Been Used? Large Language Model Prompt Membership Inference [56.20586932251531]
  We develop Prompt Detective, a statistical method to reliably determine whether a given system prompt was used by a third-party language model.<n>Our work reveals that even minor changes in system prompts manifest in distinct response distributions, enabling us to verify prompt usage with statistical significance.
  arXiv  Detail & Related papers  (2025-02-14T08:00:42Z)
• TAPO: Task-Referenced Adaptation for Prompt Optimization [18.533289140594146]
  We introduce TAPO, a multitask-aware prompt optimization framework composed of three key modules.<n>First, a task-aware metric selection module is proposed to enhance task-specific prompt generation capabilities.<n>Second, we present a multi-metrics evaluation module to jointly evaluate prompts from multiple perspectives.<n>Third, an evolution-based optimization framework is introduced for automatic prompt refinement, which improves adaptability across various tasks.
  arXiv  Detail & Related papers  (2025-01-12T02:43:59Z)
• iPrOp: Interactive Prompt Optimization for Large Language Models with a Human in the Loop [10.210078164737245]
  This paper introduces $textitiPrOp$, a novel Interactive Prompt Optimization system.<n>With human intervention in the optimization loop, $textitiPrOp$ offers users the flexibility to assess evolving prompts.
  arXiv  Detail & Related papers  (2024-12-17T08:09:15Z)
• SPRIG: Improving Large Language Model Performance by System Prompt Optimization [45.96513122345295]
  Large Language Models (LLMs) have shown impressive capabilities in many scenarios, but their performance depends on the choice of prompt. We propose SPRIG, an edit-based genetic algorithm that iteratively constructs prompts from prespecified components to maximize the model's performance in general scenarios. We evaluate the performance of system prompts on a collection of 47 different types of tasks to ensure generalizability.
  arXiv  Detail & Related papers  (2024-10-18T18:51:44Z)
• Enhancing Few-Shot Transfer Learning with Optimized Multi-Task Prompt Tuning through Modular Prompt Composition [0.0]
  Multi-task prompt tuning has garnered considerable attention for its inherent modularity and potential to enhance parameter-efficient transfer learning. This paper aims to analyze and improve the performance of multiple tasks by facilitating the transfer of knowledge between their corresponding prompts in a multi-task setting.
  arXiv  Detail & Related papers  (2024-08-23T17:01:51Z)
• QPO: Query-dependent Prompt Optimization via Multi-Loop Offline Reinforcement Learning [58.767866109043055]
  We introduce Query-dependent Prompt Optimization (QPO), which iteratively fine-tune a small pretrained language model to generate optimal prompts tailored to the input queries. We derive insights from offline prompting demonstration data, which already exists in large quantities as a by-product of benchmarking diverse prompts on open-sourced tasks. Experiments on various LLM scales and diverse NLP and math tasks demonstrate the efficacy and cost-efficiency of our method in both zero-shot and few-shot scenarios.
  arXiv  Detail & Related papers  (2024-08-20T03:06:48Z)
• Multitask Vision-Language Prompt Tuning [103.5967011236282]
  We propose multitask vision-language prompt tuning (MV) MV incorporates cross-task knowledge into prompt tuning for vision-language models. Results in 20 vision tasks demonstrate that the proposed approach outperforms all single-task baseline prompt tuning methods.
  arXiv  Detail & Related papers  (2022-11-21T18:41:44Z)
• Optimizing Interactive Systems via Data-Driven Objectives [70.3578528542663]
  We propose an approach that infers the objective directly from observed user interactions. These inferences can be made regardless of prior knowledge and across different types of user behavior. We introduce Interactive System (ISO), a novel algorithm that uses these inferred objectives for optimization.
  arXiv  Detail & Related papers  (2020-06-19T20:49:14Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.