AgentSSL：让机器学习智能体用半监督学习利用无标注数据
AgentSSL: Can MLE Agents Leverage Unlabeled Data?
这篇论文回答了一个很实际的问题：标注数据不够时，MLE 智能体还能不能自动搭好训练流程。结果显示智能体在多数设定下能超过人工基线，但选哪个程序是个大坑，最多掉 47% 性能。做自动化机器学习或数据稀缺场景的工程团队值得看看。
arXiv 论文提出 Agentic Semi-Supervised Learning（AgentSSL），把半监督学习（SSL）建模为程序搜索问题，研究 MLE 智能体在标注数据稀缺、无标注数据充足的场景下能否自动设计训练流程。
实验显示，在弱反馈和同等算力条件下，智能体生成的代码能组合并改造成熟的 SSL 技术，在 12 个设定中的 8 个超过最强人工基线，最高领先 +21.8%。但论文发现“程序选择”是关键瓶颈：若依赖智能体自身反馈指标来选程序，性能最多下降 47%。
为系统研究该瓶颈，作者构建 ProgramSelectBench，包含 16,977 个由智能体生成和人工编写的 SSL 程序，评测了无监督与主动式两类程序选择方法。论文还在真实生态学检测任务上验证 AgentSSL，指出把智能体发现的策略迁移到受控基准之外的挑战。
AgentSSL: Can MLE Agents Leverage Unlabeled Data?
Applying machine learning systems to new domains introduces bottlenecks in both labeled data acquisition and manual pipeline engineering. While recent MLE agents show promise in automating pipeline design, they have been studied almost exclusively in fully supervised settings, leaving them untested in realistic regimes where labeled data is scarce but unlabeled data is abundant. In this work, we introduce Agentic Semi-Supervised Learning (AgentSSL), framing SSL as a program search problem and studying how agents navigate this space under limited supervision. We find that even under weak feedback and matched compute, agents can synthesize code that combines and adapts established SSL techniques, exceeding the strongest human-engineered baselines in 8 of 12 settings by up to +21.8%. However, our results identify program selection as a critical bottleneck in AgentSSL, where performance drops by as much as 47% when selecting based on the agent's feedback metric. To systematically study this bottleneck, we introduce ProgramSelectBench, a large-scale benchmark with a combined total of 16,977 agent-generated and human-engineered SSL programs, and evaluate both unsupervised and active program selection methods. We further evaluate AgentSSL on real-world ecology detection tasks, highlighting the opportunities and challenges of transferring agent-discovered strategies beyond controlled benchmarks.
This report is categorized under AITOP's "Papers" desk. Within our cross-regional analytical framework, it directly maps to AI developer toolchains, coding agents, and engineering workflows. Key entities and technical topics under continuous tracking include MLE Agent, 半监督学习, ProgramSelectBench, 自动机器学习.
Source attribution: curated from arXiv cs.LG (@Ethan Lin, Akanksha Sarkar, Ziang Liu, Kristin Branson, Jennifer J. Sun).