InSTA: Towards Internet-Scale Training For Agents

Brandon Trabucco, Gunnar A Sigurdsson, Robinson Piramuthu, Ruslan Salakhutdinov

Keywords: Deep Learning, Self-Improvement

Abstract: The predominant approach for training web navigation agents is to gather human demonstrations for a set of popular websites and hand-written tasks, but it is becoming clear that human data is an inefficient resource. We develop a pipeline to facilitate internet-scale training for agents without laborious human annotations. In the first stage, an LLM annotates 150k sites with agentic tasks. In the next stage, LLM agents complete tasks and produce trajectories. In the final stage, an LLM filters trajectories by judging their success. Language models are powerful data curation tools, identifying harmful content with an accuracy of 97\%, judging successful trajectories with an accuracy of 82.6\%, and producing effective data. We train agents based on \textit{Qwen 3 1.7B} that are competitive with frontier LLMs as web agents, while being smaller and faster. Our top agent reaches a success rate of 56.9\%, outperforming the data collection policy \textit{Qwen 3 235B}, a 235 times larger \textit{Llama 4 Maverick}, and reaching 94.7\% of the performance of \textit{Gemini 2.5 Flash}. We will be releasing code, models and data that reproduce the entire pipeline on our website.

Primary Area: applications to computer vision, audio, language, and other modalities

Submission Number: 22833