Adaptive rerouting improves deep research agent decision making
RIVO: Lightweight Adaptive Rerouting for Deep Research Agents
Computation and Language
Summary
Deep Research agents try to solve complex problems by collecting and reasoning through a lot of information step by step. Sometimes they stop too soon, keep working when it’s not needed, or follow unhelpful paths. The authors of this paper created RIVO, a small add-on tool that helps these agents decide when and how to change their approach without retraining the whole system. RIVO tweaks the agent’s internal signals to steer it toward better research directions, leading to improved results on various tests.
What this means in practice
- •For ai developers: Integrate lightweight adaptive rerouting to improve complex research task performance without retraining large language models.
- •For knowledge management teams: Use adaptive rerouting tools to enhance the effectiveness of automated agents exploring large document corpora for insights.
Authors
Mingchen Li, Jiatan Huang, Hong Yu
Abstract
Deep Research agents rely on long-horizon reasoning and iterative tool use to solve complex information-seeking tasks. However, adaptive decision-making over these trajectories remains challenging, as agents may stop prematurely when further evidence is needed, continue unnecessarily after sufficient evidence has been gathered, or fail to redirect unproductive trajectories. Existing solutions either require costly model retraining and dense turn-level supervision, or rely on heuristic control rules that are not directly optimized for downstream research utility. This motivates a lightweight, utility-aware alternative for Deep Research control. We formulate adaptive rerouting in long-horizon Deep Research as a state-conditioned trajectory correction problem. Based on this view, we introduce RIVO, a lightweight plug-in framework for adaptive trajectory correction. The agent estimates its research state from lightweight signals and determines whether trajectory redirection is needed. Once triggered, the plug-in modifies the Host token logits to steer generation toward a more promising research direction. Theoretical analysis shows that, under a value-alignment condition and bounded local curvature, a sufficiently small logit intervention can locally improve expected downstream research utility. We further construct BrowseCore, a curated dataset with a paired local corpus, for efficient and controlled Deep Research experimentation. Building on this resource, the plug-in is trained from downstream outcomes, enabling it to learn corrections that improve subsequent research and final task performance. Extensive experiments across multiple Deep Research benchmarks and control settings demonstrate the effectiveness of our method, showing that lightweight logit intervention can improve long-horizon research without retraining the underlying Host model.