paper-with-me

Papers

UPA: Unsupervised Prompt Agent via Tree-Based Search and Selection

2026-01-30 · Siran Peng, Weisong Zhao, Tianyu Fu, Chenxu Zhao, Tianshuo Zhang, Haoyuan Zhang, Xiangyu Zhu, Minghui Wu, Zhen Lei arxiv

Prompt agents have recently emerged as a promising paradigm for automated prompt optimization, framing prompt discovery as a sequential decision-making problem over a structured prompt space. While this formulation enables the use of advanced planning algorithms, these methods typically assume access to supervised reward signals, which are often unavailable in practical scenarios. In this work, we propose UPA, an Unsupervised Prompt Agent that realizes structured search and selection without relying on ground-truth (GT) rewards. Specifically, during search, UPA iteratively constructs an evolving tree structure to navigate the prompt space, guided by fine-grained and position-debiased pairwise comparisons from Large Language Models (LLMs). Crucially, as these local comparisons do not inherently yield a consistent global scale, we decouple systematic prompt exploration from final selection, introducing a two-stage framework grounded in the Bradley-Terry-Luce (BTL) model. This framework first performs path-wise Bayesian aggregation of local comparisons to filter candidates under uncertainty, followed by global tournament-style comparisons to infer latent prompt quality and identify the optimal prompt. Experiments across multiple tasks demonstrate that UPA consistently outperforms existing prompt optimization methods, showing that agent-style optimization can remain highly effective even in unsupervised settings.

📄 PDF Abstract BibTeX arXiv:2601.23273

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Maximizing Rollout Informativeness under a Fixed Budget: A Submodular View of Tree Search for Tool-Use Agentic Reinforcement Learning

2026-05-06 · Yuelin Hu, Zhenbo Yu, Zhengxue Cheng, Wei Liu 외 arxiv

We formalize Rollout Informativeness under a Fixed Budget (RIFB) as the expected non-vanishing policy-gradient mass that a tool-use rollout set injects into Group Relative Policy Optimization (GRPO). We prove that any bu…

Reinforcement Learning

RoT: Enhancing Large Language Models with Reflection on Search Trees

2024-04-08 · Wenyang Hui, Kewei Tu

Large language models (LLMs) have demonstrated impressive capability in reasoning and planning when integrated with tree-search-based prompting methods. However, since these methods ignore the previous search experiences…

TreePrompt: Leveraging Hierarchical Few-Shot Example Selection for Improved English-Persian and English-German Translation

2025-10-04 · Ramtin Kakavand, Ebrahim Ansari arxiv

Large Language Models (LLMs) have consistently demonstrated strong performance in machine translation, especially when guided by high-quality prompts. Few-shot prompting is an effective technique to improve translation q…

Machine Translation

AgentVigil: Generic Black-Box Red-teaming for Indirect Prompt Injection against LLM Agents

2025-05-09 · Zhun Wang, Vincent Siu, Zhe Ye, Tianneng Shi 외

The strong planning and reasoning capabilities of Large Language Models (LLMs) have fostered the development of agent-based systems capable of leveraging external tools and interacting with increasingly complex environme…

NavigateRed Teaming

TabTracer: Monte Carlo Tree Search for Complex Table Reasoning with Large Language Models

2026-02-15 · Zhizhao Luo, Zhaojing Luo, Meihui Zhang, Rui Mao arxiv

Large language models (LLMs) have emerged as powerful tools for natural language table reasoning, where there are two main categories of methods. Prompt-based approaches rely on language-only inference or one-pass progra…