C-MOP: Integrating Momentum and Boundary-Aware Clustering for Enhanced Prompt Evolution

• URL: http://arxiv.org/abs/2602.10874v1
• Date: Wed, 11 Feb 2026 14:04:47 GMT
• Title: C-MOP: Integrating Momentum and Boundary-Aware Clustering for Enhanced Prompt Evolution
• Authors: Binwei Yan, Yifei Fu, Mingjian Zhu, Hanting Chen, Mingxuan Yuan, Yunhe Wang, Hailin Hu,
• Abstract summary: We propose C-MOP, a framework that stabilizes optimization via Boundary-Aware Contrastive Sampling (BACS) and Momentum-Guided Semantic Clustering (MGSC)<n>C-MOP consistently outperforms SOTA baselines like PromptWizard and ProTeGi, yielding average gains of 1.58% and 3.35%.
• Score: 37.02233725807037
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Automatic prompt optimization is a promising direction to boost the performance of Large Language Models (LLMs). However, existing methods often suffer from noisy and conflicting update signals. In this research, we propose C-MOP (Cluster-based Momentum Optimized Prompting), a framework that stabilizes optimization via Boundary-Aware Contrastive Sampling (BACS) and Momentum-Guided Semantic Clustering (MGSC). Specifically, BACS utilizes batch-level information to mine tripartite features--Hard Negatives, Anchors, and Boundary Pairs--to precisely characterize the typical representation and decision boundaries of positive and negative prompt samples. To resolve semantic conflicts, MGSC introduces a textual momentum mechanism with temporal decay that distills persistent consensus from fluctuating gradients across iterations. Extensive experiments demonstrate that C-MOP consistently outperforms SOTA baselines like PromptWizard and ProTeGi, yielding average gains of 1.58% and 3.35%. Notably, C-MOP enables a general LLM with 3B activated parameters to surpass a 70B domain-specific dense LLM, highlighting its effectiveness in driving precise prompt evolution. The code is available at https://github.com/huawei-noah/noah-research/tree/master/C-MOP.

Related papers

• Multi-Paradigm Collaborative Adversarial Attack Against Multi-Modal Large Language Models [67.45032003041399]
  We propose a novel Multi-Paradigm Collaborative Attack (MPCAttack) framework to boost the transferability of adversarial examples against MLLMs.<n>MPCO adaptively balances the importance of different paradigm representations and guides the global optimisation.<n>Our solution consistently outperforms state-of-the-art methods in both targeted and untargeted attacks on open-source and closed-source MLLMs.
  arXiv  Detail & Related papers  (2026-03-05T06:01:26Z)
• Empowering Contrastive Federated Sequential Recommendation with LLMs [13.016120865205258]
  Federated sequential recommendation (FedSeqRec) aims to perform next-item prediction while keeping user data decentralised.<n>We propose textbfLUMOS, a parameter-isolated FedSeqRec architecture that integrates large language models (LLMs) as emphlocal semantic generators.
  arXiv  Detail & Related papers  (2026-02-10T00:47:43Z)
• Multi-Perspective Subimage CLIP with Keyword Guidance for Remote Sensing Image-Text Retrieval [18.55080473948215]
  MPS-CLIP is a parameter-efficient framework designed to shift the retrieval paradigm from global matching to keyword-guided fine-grained alignment.<n>Experiments on the RSICD and RSITMD benchmarks demonstrate that MPS-CLIP achieves state-of-the-art performance with 35.18% and 48.40% mean Recall.
  arXiv  Detail & Related papers  (2026-01-26T06:16:53Z)
• GAP-Net: Calibrating User Intent via Gated Adaptive Progressive Learning for CTR Prediction [0.6372261626436676]
  GAP-Net is a unified framework establishing a "Triple Gating" architecture to progressively refine information from micro-level features to macro-level views.<n>It achieves substantial improvements over state-of-the-art baselines, exhibiting superior robustness against interaction noise and intent drift.
  arXiv  Detail & Related papers  (2026-01-12T15:01:12Z)
• PROMISE: Process Reward Models Unlock Test-Time Scaling Laws in Generative Recommendations [52.67948063133533]
  Generative Recommendation has emerged as a promising paradigm, reformulating recommendation as a sequence-to-sequence generation task over hierarchical Semantic IDs.<n>Existing methods suffer from a critical issue we term Semantic Drift, where errors in early, high-level tokens irreversibly divert the generation trajectory into irrelevant semantic subspaces.<n>We propose Promise, a novel framework that integrates dense, step-by-step verification into generative models.
  arXiv  Detail & Related papers  (2026-01-08T07:38:46Z)
• GrAInS: Gradient-based Attribution for Inference-Time Steering of LLMs and VLMs [56.93583799109029]
  GrAInS is an inference-time steering approach that operates across both language-only and vision-language models and tasks.<n>During inference, GrAInS hidden activations at transformer layers guided by token-level attribution signals, and normalizes activations to preserve representational scale.<n>It consistently outperforms both fine-tuning and existing steering baselines.
  arXiv  Detail & Related papers  (2025-07-24T02:34:13Z)
• Pangu Embedded: An Efficient Dual-system LLM Reasoner with Metacognition [95.54406667705999]
  Pangu Embedded is an efficient Large Language Model (LLM) reasoner developed on Ascend Neural Processing Units (NPUs)<n>It addresses the significant computational costs and inference latency challenges prevalent in existing reasoning-optimized LLMs.<n>It delivers rapid responses and state-of-the-art reasoning quality within a single, unified model architecture.
  arXiv  Detail & Related papers  (2025-05-28T14:03:02Z)
• Introducing MAPO: Momentum-Aided Gradient Descent Prompt Optimization [2.750784330885499]
  Building on ProTeGi, MAPO uses positive natural language "gradients" and a momentum-based extension to refine prompts effectively.<n>MAPO achieves faster convergence time with fewer API calls and higher F1 scores than ProTeGi.
  arXiv  Detail & Related papers  (2024-10-25T11:58:12Z)
• Unleashing Network Potentials for Semantic Scene Completion [50.95486458217653]
  This paper proposes a novel SSC framework - Adrial Modality Modulation Network (AMMNet) AMMNet introduces two core modules: a cross-modal modulation enabling the interdependence of gradient flows between modalities, and a customized adversarial training scheme leveraging dynamic gradient competition. Extensive experimental results demonstrate that AMMNet outperforms state-of-the-art SSC methods by a large margin.
  arXiv  Detail & Related papers  (2024-03-12T11:48:49Z)
• Faster Learning of Temporal Action Proposal via Sparse Multilevel Boundary Generator [9.038216757761955]
  Temporal action localization in videos presents significant challenges in the field of computer vision. We propose a novel framework, Sparse Multilevel Boundary Generator (SMBG), which enhances the boundary-sensitive method with boundary classification and action completeness regression. Our method is evaluated on two popular benchmarks, ActivityNet-1.3 and THUMOS14, and is shown to achieve state-of-the-art performance, with a better inference speed (2.47xBSN++, 2.12xDBG)
  arXiv  Detail & Related papers  (2023-03-06T14:26:56Z)

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.