An AI agentAn AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.Full definition → can prepare useful resources before it knows which tasks it will receive. In this research, a studying agent explores available documents and tools within a budget, then leaves behind resources such as indexes, scripts, or procedural guidance for an unchanged task-solving system.
Across six benchmarks, the authors report that these resources reduced the repeated attempts needed to reach a given score. This moves some computation into reusable preparation. Giving the studying agent a larger budget did not reliably improve results, and a fixed document-processing method remained strongest on the largest document collection tested.
Terms in this piece · Glossary
AI agent — An AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.
LLM — A large language model — the neural network behind tools like Claude and ChatGPT, trained on huge amounts of text to predict what comes next.
articleAgentic Data Engineering for LLMs: System Design and a Controlled Empirical StudyRuomeng Ding, Qianli Shen, Zhaoyang Han, Meixin Chen, Daoyuan Chen, Yaliang Li
articleSubagents vs Agent Skills: Executing Reusable Knowledge for Long-Horizon Agentic TasksWasu Top Piriyakulkij, Rachel Lawrence, Alicia Curth, Sushrut Karmalkar, Niranjani Prasad
articleScientific Agent Skills: A Library of Procedural Knowledge for Research AgentsTimothy Kassis, Vinayak Agarwal, Yuhuan He, Darshil Patel, Aubrey M. Brueckner