AI Insight
This study introduces Speculative Rollback Correction (SRC), a new training method for web-based AI agents that learn from expert demonstrations. The approach allows agents to execute short sequences of actions before expert review, with corrections applied only when the agent makes harmful mistakes, rather than at every step or only after complete failure. Testing on WebArena-Infinity generated 977 successful trajectories and 9,183 training examples, improving the balance between learning efficiency and maintaining diverse solution strategies.
Why it matters
This method could significantly improve the training of AI agents for web-based tasks like automated form filling, online shopping, or web navigation by reducing over-reliance on rigid, single-path solutions. The approach addresses a key challenge in autonomous agent development by finding an optimal middle ground between too much and too little expert guidance.
Understand the Science
⚠️ Preprint – Noch nicht peer-reviewed
Dieser Artikel wurde noch nicht von unabhängigen Experten begutachtet. Die Ergebnisse sind vorläufig und sollten mit Vorsicht interpretiert werden.
Abstract: Training interactive web agents through imitation learning from expert trajectories has emerged as a highly effective approach. However, determining the optimal timing for expert intervention presents a critical challenge in this context. Delayed intervention often leads to the accumulation of early-stage errors, pushing the page state into an irrecoverable regime. Conversely, premature or excessive intervention causes the agent to become overly reliant on expert policies, trapping the model in local optima characterized by a single, rigid trajectory. We propose Speculative Rollback Correction (SRC), a branch-level imitation framework for resettable agent environments. Instead of requesting teacher labels at every visited state or correcting only after a completed trajectory, SRC uses fixed-horizon branch review: the student executes a short speculative segment before teacher review, and the teacher localizes the first harmful deviation only when local progress breaks. Rollback preserves useful prefixes, while successful rollouts are filtered by a hard verifier and retained in a lightweight quality-diversity archive. The resulting data supports next-action supervised fine-tuning on both localized corrections and verifier-passing trajectories. On WebArena-Infinity, SRC collects 977 verifier-passing trajectories and 9,183 next-action examples; fixed-horizon review improves the recovery-versus-query tradeoff over step-level review while retaining verifier-passing solution variants. Code is available at https://github.com/LongkunHao/SRC_gui_agent.
Source: Speculative Rollback Correction for Quality-Diverse Web Agent Imitation