{"slug": "agentssl-can-mle-agents-leverage-unlabeled-data", "title": "AgentSSL: Can MLE Agents Leverage Unlabeled Data?", "summary": "A paper submitted to arXiv on 12 Nov 2026 introduces Agentic Semi-Supervised Learning (AgentSSL), framing semi-supervised learning as a program search problem for machine-learning engineering agents. Under weak feedback and matched compute, the agents synthesized code combining established SSL techniques that beat the strongest human-engineered baselines in 8 of 12 settings by up to 21.8%, but program selection proved a critical bottleneck, with performance dropping by as much as 47% when selecting on the agent's feedback metric. The authors released ProgramSelectBench, a benchmark of 16,977 agent-generated and human-engineered SSL programs, and also evaluated AgentSSL on real-world ecology detection tasks.", "body_md": "# Computer Science > Artificial Intelligence\n\n  [Submitted on 12 Nov 2026]\n\n# Title:AgentSSL: Can MLE Agents Leverage Unlabeled Data?\n\n[View PDF](http://arxiv.org/pdf/2611.17323v1)\n\n[HTML (experimental)](https://arxiv.org/html/2611.17323v1)\n\nAbstract: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.\n    \n\n### Additional Features\n\n### References & Citations\n\nLoading...\n\n# Bibliographic and Citation Tools\n\nBibliographic Explorer \n\n*(*[What is the Explorer?](https://info.arxiv.org/labs/showcase.html#arxiv-bibliographic-explorer))\nConnected Papers \n\n*(*[What is Connected Papers?](https://www.connectedpapers.com/about))\nLitmaps \n\n*(*[What is Litmaps?](https://www.litmaps.co/))\nscite Smart Citations \n\n*(*[What are Smart Citations?](https://www.scite.ai/))\n# Code, Data and Media Associated with this Article\n\nalphaXiv \n\n*(*[What is alphaXiv?](https://alphaxiv.org/))\nCatalyzeX Code Finder for Papers \n\n*(*[What is CatalyzeX?](https://www.catalyzex.com))\nDagsHub \n\n*(*[What is DagsHub?](https://dagshub.com/))\nGotit.pub \n\n*(*[What is GotitPub?](http://gotit.pub/faq))\nHugging Face \n\n*(*[What is Huggingface?](https://huggingface.co/huggingface))\nScienceCast \n\n*(*[What is ScienceCast?](https://sciencecast.org/welcome))\n# Demos\n\n# Recommenders and Search Tools\n\nInfluence Flower \n\n*(*[What are Influence Flowers?](https://influencemap.cmlab.dev/))\nCORE Recommender \n\n*(*[What is CORE?](https://core.ac.uk/services/recommender))\n# arXivLabs: experimental projects with community collaborators\n\narXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.\n\nBoth individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.\n\nHave an idea for a project that will add value for arXiv's community? [**Learn more about arXivLabs**](https://info.arxiv.org/labs/index.html).", "url": "https://wpnews.pro/news/agentssl-can-mle-agents-leverage-unlabeled-data", "canonical_source": "http://arxiv.org/abs/2611.17323v1", "published_at": "2026-11-13 15:44:32+00:00", "updated_at": "2026-11-13 16:22:09.532310+00:00", "lang": "en", "topics": ["machine-learning", "ai-agents", "ai-research", "artificial-intelligence"], "entities": ["AgentSSL", "ProgramSelectBench", "arXiv"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/agentssl-can-mle-agents-leverage-unlabeled-data", "markdown": "https://wpnews.pro/news/agentssl-can-mle-agents-leverage-unlabeled-data.md", "text": "https://wpnews.pro/news/agentssl-can-mle-agents-leverage-unlabeled-data.txt", "jsonld": "https://wpnews.pro/news/agentssl-can-mle-agents-leverage-unlabeled-data.jsonld"}}