Equitable System-Prompt Selection via Constrained Mixed-Strategy GroupDRO

• URL: http://arxiv.org/abs/2608.04339v1
• Date: Wed, 05 Aug 2026 01:35:08 GMT
• Title: Equitable System-Prompt Selection via Constrained Mixed-Strategy GroupDRO
• Abstract summary: System prompts are widely employed to steer response behavior, but they are typically optimized for average-case quality.<n>We formulate a constrained mixed-strategy GroupDRO framework for system-prompt selection.<n>Our framework assigns weights to system prompts in an existing pool to minimize the worst-case information-quality loss.
• Score: 14.116598550478722
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Large language models are increasingly used for information seeking, yet semantically equivalent questions phrased in different ways can receive answers of considerably different quality. System prompts are widely employed to steer response behavior, but they are typically optimized for average-case quality, so some question phrasings may still receive incomplete or low-quality answers. To address this, we formulate a constrained mixed-strategy GroupDRO framework for system-prompt selection. Instead of optimizing the system-prompt text, the framework assigns weights to system prompts in an existing pool to minimize the worst-case information-quality loss across evaluation metrics and groups, while constraining the mean loss to stay close to that of average-based selection. Because pool generation and selection are decoupled, the method applies to any system-prompt pool and can leverage an ensemble of complementary system prompts rather than a single one. Across five LLMs on two bilingual medical and consumer-finance benchmarks, the constrained method reduces the Overall Mean, Worst 25% Mean, and Worst by 13.1%, 13.2%, and 13.7% on average relative to no mitigation while keeping overall quality close to Average selection. Its multi-prompt weights reveal complementarity across metric-group pairs. Code and data are available at https://github.com/Rainxu09/equitable-system-prompt-selection.

Related papers

• Pre-retrieval Query Clustering for Adaptive Top-k Document Retrieval in RAG Systems [7.440012812118169]
  RAG systems commonly retrieve a fixed number of documents to ground generation.<n>simple queries suffer over-retrieval (adding noise and cost) while complex queries are under-retrieved.<n>We propose a practical, general framework for query-adaptive retrieval depth.
  arXiv  Detail & Related papers  (2026-09-11T19:57:12Z)
• SSPO: Structure-Aware Similarity-Weighted Preference Optimization for Neural Combinatorial Optimization [7.503926194234302]
  Preference-optimization methods anchor on the single best solution and discard fine-grained quality and structural signal from all other peers.<n>We propose SSPO (Structure-Aware Similarity-Weighted Preference Optimization), which scores all $B$ sampled solutions jointly through a dissimilarity-weighted leave-one-out baseline.
  arXiv  Detail & Related papers  (2026-08-12T16:04:54Z)
• F-GRPO: Factorized Group-Relative Policy Optimization for Unified Candidate Generation and Ranking [79.49893545611779]
  Large Language Models (LLMs) can generate a subset and order it within a single autoregressive pass.<n>This flexibility introduces a new optimization challenge: the model must search an output space while receiving utility feedback only after the full ranked list is generated.<n>This credit assignment gap makes end-to-end optimization unstable and sample-inefficient.<n>We propose a unified framework that performs both within a single autoregressive rollout and optimize them end-to-end via factorized group-relative policy optimization.
  arXiv  Detail & Related papers  (2026-05-13T04:52:33Z)
• MASS-DPO: Multi-negative Active Sample Selection for Direct Policy Optimization [66.81594613265833]
  Multi-negative preference optimization under the Plackett--Luce (PL) model extends Direct Preference Optimization (DPO)<n>We introduce MASS-DPO, a multi-negative active sample selection method that derives a PL-specific Fisher-information objective for selecting compact, informative negative subsets within each prompt.<n> MASS-DPO consistently exceeds or matches existing methods in accuracy, improves Recall/NDCG and margin-based optimization dynamics, and delivers stronger alignment with substantially fewer negatives.
  arXiv  Detail & Related papers  (2026-05-11T16:18:08Z)
• $p1$: Better Prompt Optimization with Fewer Prompts [49.20082664169319]
  We show that prompt optimization succeeds when variance among system prompts is large, but fails when variance among responses dominates the variance of the system prompts.<n>We propose $p1$, a simple user prompt filtering method that selects a small subset of user prompts with high variance across candidate system prompts.
  arXiv  Detail & Related papers  (2026-04-09T22:31:15Z)
• Evaluation of Multi- and Single-objective Learning Algorithms for Imbalanced Data [0.0]
  Machine learning tasks aim to find models that work well not for a single, but for a group of criteria, often opposing ones.<n>One solution is to propose an aggregate learning criterion and reduce the multi-objective learning task to a single-criteria optimization problem.<n>This article proposes a new, reliable way of evaluating algorithms based on multi-objective algorithms with methods that return single solutions.
  arXiv  Detail & Related papers  (2025-11-15T12:54:17Z)
• MUSS: Multilevel Subset Selection for Relevance and Diversity [4.8254343133177295]
  In recommender systems, one is interested in selecting relevant items, while providing a diversified recommendation.<n>We present a novel theoretical approach for analyzing this type of problems, and show that our method achieves a constant factor approximation of the optimal objective.
  arXiv  Detail & Related papers  (2025-03-14T06:37:17Z)
• MICINet: Multi-Level Inter-Class Confusing Information Removal for Reliable Multimodal Classification [57.08108545219043]
  A reliable multimodal classification method dubbed Multi-Level Inter-Class Confusing Information Removal Network (MICINet) is proposed.<n>MICINet achieves the reliable removal of both types of noise by unifying them into the concept of Inter-class Confusing Information (textitICI) and eliminating it at both global and individual levels.<n>Experiments on four datasets demonstrate that MICINet outperforms other state-of-the-art reliable multimodal classification methods under various noise conditions.
  arXiv  Detail & Related papers  (2025-02-27T01:33:28Z)
• Adaptive-RAG: Learning to Adapt Retrieval-Augmented Large Language Models through Question Complexity [59.57065228857247]
  Retrieval-augmented Large Language Models (LLMs) have emerged as a promising approach to enhancing response accuracy in several tasks, such as Question-Answering (QA) We propose a novel adaptive QA framework, that can dynamically select the most suitable strategy for (retrieval-augmented) LLMs based on the query complexity. We validate our model on a set of open-domain QA datasets, covering multiple query complexities, and show that ours enhances the overall efficiency and accuracy of QA systems.
  arXiv  Detail & Related papers  (2024-03-21T13:52:30Z)
• Minimizing Age of Information for Mobile Edge Computing Systems: A Nested Index Approach [11.998034941401814]
  Mobile edge computation (MEC) provides an efficient approach to achieving real-time applications that are sensitive to information freshness. In this paper, we consider multiple users offloading tasks to heterogeneous edge servers in a MEC system. Our algorithm leads to an optimality gap reduction of up to 40%, compared to benchmarks.
  arXiv  Detail & Related papers  (2023-07-03T21:47:21Z)
• Optimal Clustering from Noisy Binary Feedback [75.17453757892152]
  We study the problem of clustering a set of items from binary user feedback. We devise an algorithm with a minimal cluster recovery error rate. For adaptive selection, we develop an algorithm inspired by the derivation of the information-theoretical error lower bounds.
  arXiv  Detail & Related papers  (2019-10-14T09:18:26Z)

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.