Training Prompt Matters: State-Adaptive Optimization for Robust Fine-Tuning

• URL: http://arxiv.org/abs/2606.01967v1
• Date: Mon, 01 Jun 2026 09:31:28 GMT
• Title: Training Prompt Matters: State-Adaptive Optimization for Robust Fine-Tuning
• Abstract summary: We show that paraphrased prompts induce drastically different cross-task impacts regarding forgetting and generalization.<n>These impacts are positively correlated across tasks, indicating the existence of superior prompts that consistently yield better performance.<n>We introduce State-Adaptive Prompt Optimization (SAPO), a lightweight yet effective training strategy that shifts task formulation from a static input to a dynamic, state-adaptive variable.
• Score: 14.317667911880482
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: While prompt engineering is instrumental in maximizing the capabilities of Large Language Models (LLMs) during inference, the role of prompts during training remains critically underexplored. Prevailing fine-tuning paradigms typically treat training prompts as mere surface forms, assuming that semantically equivalent instructions yield identical learning outcomes. However, we reveal that this equivalence is deceptive: while paraphrased prompts often lead to comparable in-task performance, they induce drastically different cross-task impacts regarding catastrophic forgetting and generalization. Crucially, these impacts are positively correlated across tasks, indicating the existence of superior prompts that consistently yield better performance. Furthermore, we discover that these superior prompts can be robustly identified by task loss prior to learning. Leveraging these insights, we introduce State-Adaptive Prompt Optimization (SAPO), a lightweight yet effective training strategy that shifts task formulation from a static input to a dynamic, state-adaptive variable. Comprehensive experiments on diverse benchmarks confirm its effectiveness, which significantly mitigates forgetting while improving generalization, achieving substantial performance gains over state-of-the-art methods. These results provide insights into how training prompts shape learning dynamics and offer a practical recipe for robust fine-tuning. Our code is available at https://github.com/Eric8932/SAPO.

Related papers

• CVPO: Enhancing LLM Reinforcement Learning Reasoning via Value-Variance Adaptation and Dynamic Curriculum Learning [5.170275696099583]
  Reinforcement learning (RL) has emerged as an effective method for enhancing the reasoning capabilities of large language models (LLMs)<n>Existing methods suffer from insufficient precision in feedback on generated answer trajectories and exhibit the phenomenon of problem difficulty drift.<n>We propose CVPO - Curriculum-guided Value-Variance Policy Optimization.
  arXiv  Detail & Related papers  (2026-08-04T03:30:59Z)
• Dynamic Prompt Fusion for Multi-Task and Cross-Domain Adaptation in LLMs [2.852258765983155]
  This study introduces a unified multi-task learning framework with dynamic prompt scheduling mechanism.<n>It enhances the model's ability to capture semantic differences across tasks.<n>It incorporates an automatic learning strategy for scheduling weights, which effectively mitigates task interference and negative transfer.
  arXiv  Detail & Related papers  (2025-09-09T23:42:16Z)
• Underlying Semantic Diffusion for Effective and Efficient In-Context Learning [113.4003355229632]
  Underlying Semantic Diffusion (US-Diffusion) is an enhanced diffusion model that boosts underlying semantics learning, computational efficiency, and in-context learning capabilities.<n>We present a Feedback-Aided Learning (FAL) framework, which leverages feedback signals to guide the model in capturing semantic details.<n>We also propose a plug-and-play Efficient Sampling Strategy (ESS) for dense sampling at time steps with high-noise levels.
  arXiv  Detail & Related papers  (2025-03-06T03:06:22Z)
• PAFT: Prompt-Agnostic Fine-Tuning [24.209557567599393]
  Fine-tuning large language models (LLMs) often causes overfitting to specific prompt wording, where minor phrasing variations drastically reduce performance.<n>We propose Prompt-Agnostic Fine-Tuning (PAFT), a method that enhances robustness through dynamic prompt variation during training.
  arXiv  Detail & Related papers  (2025-02-18T13:46:47Z)
• Prompt-Tuning Bandits: Enabling Few-Shot Generalization for Efficient Multi-Task Offline RL [2.6731152954002924]
  We propose a lightweight, inference-time, bandit-based prompt-tuning framework.<n>The bandit explores and optimize trajectory prompt selection to enhance task performance, while avoiding costly fine-tuning of the transformer backbone.<n>Our experiments indicate not only clear performance gains due to bandit-based prompt-tuning, but also better sample complexity, scalability, and prompt space exploration.
  arXiv  Detail & Related papers  (2025-02-10T11:20:10Z)
• Think Beyond Size: Adaptive Prompting for More Effective Reasoning [0.0]
  We introduce Adaptive Prompting, a dynamic and iterative framework designed to enhance reasoning by incorporating real-time adjustments to prompt structures and validation mechanisms.<n>Results demonstrate that Adaptive Prompting significantly improves performance on diverse reasoning benchmarks, including arithmetic reasoning (GSM8K, MultiArithm), logical reasoning and commonsense tasks.<n>Our approach enables smaller models to achieve competitive performance with larger counterparts, such as GPT-4, while maintaining computational efficiency.
  arXiv  Detail & Related papers  (2024-10-10T17:14:36Z)
• Denoising Pre-Training and Customized Prompt Learning for Efficient Multi-Behavior Sequential Recommendation [69.60321475454843]
  We propose DPCPL, the first pre-training and prompt-tuning paradigm tailored for Multi-Behavior Sequential Recommendation. In the pre-training stage, we propose a novel Efficient Behavior Miner (EBM) to filter out the noise at multiple time scales. Subsequently, we propose to tune the pre-trained model in a highly efficient manner with the proposed Customized Prompt Learning (CPL) module.
  arXiv  Detail & Related papers  (2024-08-21T06:48:38Z)
• DETAIL: Task DEmonsTration Attribution for Interpretable In-context Learning [75.68193159293425]
  In-context learning (ICL) allows transformer-based language models to learn a specific task with a few "task demonstrations" without updating their parameters.<n>We propose an influence function-based attribution technique, DETAIL, that addresses the specific characteristics of ICL.<n>We experimentally prove the wide applicability of DETAIL by showing our attribution scores obtained on white-box models are transferable to black-box models in improving model performance.
  arXiv  Detail & Related papers  (2024-05-22T15:52:52Z)
• Active Instruction Tuning: Improving Cross-Task Generalization by Training on Prompt Sensitive Tasks [101.40633115037983]
  Instruction tuning (IT) achieves impressive zero-shot generalization results by training large language models (LLMs) on a massive amount of diverse tasks with instructions. How to select new tasks to improve the performance and generalizability of IT models remains an open question. We propose active instruction tuning based on prompt uncertainty, a novel framework to identify informative tasks, and then actively tune the models on the selected tasks.
  arXiv  Detail & Related papers  (2023-11-01T04:40:05Z)
• Hierarchical Decomposition of Prompt-Based Continual Learning: Rethinking Obscured Sub-optimality [55.88910947643436]
  Self-supervised pre-training is essential for handling vast quantities of unlabeled data in practice. HiDe-Prompt is an innovative approach that explicitly optimize the hierarchical components with an ensemble of task-specific prompts and statistics. Our experiments demonstrate the superior performance of HiDe-Prompt and its robustness to pre-training paradigms in continual learning.
  arXiv  Detail & Related papers  (2023-10-11T06:51:46Z)
• When Prompt-based Incremental Learning Does Not Meet Strong Pretraining [36.0889029038102]
  In this work, we develop a learnable Adaptive Prompt Generator (APG) The key is to unify the prompt retrieval and prompt learning processes into a learnable prompt generator. Our method significantly outperforms advanced methods in exemplar-free incremental learning without (strong) pretraining.
  arXiv  Detail & Related papers  (2023-08-21T03:33:21Z)
• Fairness-guided Few-shot Prompting for Large Language Models [93.05624064699965]
  In-context learning can suffer from high instability due to variations in training examples, example order, and prompt formats. We introduce a metric to evaluate the predictive bias of a fixed prompt against labels or a given attributes. We propose a novel search strategy based on the greedy search to identify the near-optimal prompt for improving the performance of in-context learning.
  arXiv  Detail & Related papers  (2023-03-23T12:28:25Z)
• Dynamic Prompting: A Unified Framework for Prompt Tuning [33.175097465669374]
  We present a unified dynamic prompt (DP) tuning strategy that dynamically determines different factors of prompts based on specific tasks and instances. Experimental results underscore the significant performance improvement achieved by dynamic prompt tuning across a wide range of tasks. We establish the universal applicability of our approach under full-data, few-shot, and multitask scenarios.
  arXiv  Detail & Related papers  (2023-03-06T06:04:46Z)

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.