对比搜索代理训练中检索、奖励与协议的影响，发现数据覆盖比算法更重要。
Retrieval, Reward, and Training Protocols: What Matters in Training Search Agents?
- 控制变量实验，分离检索、奖励与训练协议三因素影响。
- 修正维基百科数据覆盖问题带来的提升，超过不同算法差异。
- 结果表明简单结果奖励优于复杂过程奖励，适合实际部署。
由大语言模型驱动的搜索代理可自主分解查询、检索信息并合成答案，实现多步推理。然而，训练方法快速发展，却缺乏可控对比：现有研究在检索语料、奖励设计和训练协议上存在差异，难以判断真正推动性能提升的因素。本文开展受控实证研究，分离出三个被忽视的训练维度。首先，发现广泛使用的Wikipedia 2018语料存在严重数据覆盖率问题，仅修正该问题即可带来显著提升，超过不同训练算法间的差异。其次，在三种基础模型上系统比较基于结果和基于过程的奖励方式，发现最简单的基于结果的奖励在多数场景下表现相当或更优，而过程级信用分配可能导致行为过度矫正。第三，分析训练数据多样性、离策略数据利用及搜索预算扩展性，提炼出训练高效搜索代理的实用指南。代码已开源：https://github.com/YiboZhao624/SearchAgentReview。
原文摘要 · Abstract (English)
Search agents powered by large language models can autonomously decompose queries, retrieve information, and synthesize answers through multi-step reasoning. However, the rapid growth of training methods has outpaced controlled comparison: existing works differ in retrieval corpora, reward designs, and training protocols, making it unclear what actually drives improvements. We present a controlled empirical study that isolates three under-explored dimensions of search agent training. First, we identify a critical data-coverage issue in the widely used Wikipedia 2018 corpus and show that correcting it alone yields larger gains than the differences between training algorithms. Second, we systematically compare outcome-based and process-based reward methods across three base models, finding that the simplest outcome-based approach achieves competitive or superior performance in most settings, and that process-level credit assignment can over-correct agent behavior. Third, we analyze training data diversity, off-policy data utilization, and search budget scaling, distilling practical guidelines for training effective search agents. Our code is available at https://github.com/YiboZhao624/SearchAgentReview.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。