Weak-to-Strong Generalization with Failure Trajectories: A Tree-based Approach to Elicit Optimal Policy in Strong Models
This work addresses the challenge of efficiently training strong AI models for complex tasks, offering a scalable method that could benefit AI development in interactive environments, though it appears incremental by building on existing weak-to-strong generalization concepts.
The paper tackles the problem of eliciting the full capabilities of strong models in complex interactive decision-making environments by extending weak-to-strong generalization beyond simple tasks, using failure trajectories and trajectory trees with Monte Carlo Tree Search, resulting in substantial improvements in reasoning and decision-making across diverse domains.
Weak-to-Strong generalization (W2SG) is a new trend to elicit the full capabilities of a strong model with supervision from a weak model. While existing W2SG studies focus on simple tasks like binary classification, we extend this paradigm to complex interactive decision-making environments. Specifically, we fine-tune a strong model with trajectories of intermediate actions generated by a weak model. Motivated by the human learning process, we propose to generalize not only success knowledge but also failure experience so that the strong model can learn from failed trajectories accumulated by weak models. To effectively and efficiently elicit the potential of strong agents, we further construct ``trajectory trees," a hierarchical representation that organizes weak model-generated action trajectories, coupled with Monte Carlo Tree Search (MCTS) to optimize the strong model. Through theoretical analysis, we provide formal guarantees for the effectiveness of our method in improving W2SG performance. Our empirical evaluations demonstrate substantial improvements in reasoning and decision-making capabilities across diverse task domains, validating the scalability and robustness of our proposed framework.