Bifrost: Steering Strategic Trajectories to Bridge Contextual Gaps for Self-Improving Agents

2026-02-05

AI Summary

This work addresses the challenge of reusing historical successful trajectories when autonomous agents transfer across tasks with significant contextual shifts—a setting where existing methods either suffer from limited effectiveness or require costly fine-tuning. The study reveals, for the first time, a parallel shift relationship between context and trajectory in the latent space. Building on this insight, it proposes a training-free trajectory adaptation mechanism that represents trajectories via the agent’s hidden states and leverages contextual differences to guide and align trajectories in latent space. Evaluated across multiple benchmarks, the approach substantially outperforms current trajectory reuse and fine-tuning strategies, demonstrating that agents can efficiently repurpose past experiences even under substantial contextual changes.

Technology Category

• Multiagent Systems: Adversarial Agents
• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Search and Optimization: Learning to Search

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

Autonomous agents excel in self-improvement through reflection and iterative refinement, which reuse successful task trajectories as in-context examples to assist subsequent reasoning. However, shifting across tasks often introduces a context mismatch. Hence, existing approaches either discard the trajectories or manipulate them using heuristics, leading to a non-negligible fine-tuning cost or unguaranteed performance. To bridge this gap, we reveal a context-trajectory correlation, where shifts of context are highly parallel with shifts of trajectory. Based on this finding, we propose BrIdge contextual gap FoR imprOvised trajectory STeering (Bifrost), a training-free method that leverages context differences to precisely guide the adaptation of previously solved trajectories towards the target task, mitigating the misalignment caused by context shifts. Our trajectory adaptation is conducted at the representation level using agent hidden states, ensuring trajectory transformation accurately aligns with the target context in a shared space. Across diverse benchmarks, Bifrost consistently outperforms existing trajectory reuse and finetuned self-improvement methods, demonstrating that agents can effectively leverage past experiences despite substantial context shifts.

Problem

• context shift
• trajectory reuse
• self-improving agents
• context mismatch
• autonomous agents

Innovation

• context-trajectory correlation
• trajectory steering
• training-free adaptation
• representation alignment
• self-improving agents