Masking Stale Observations Helps Search Agents -- Until It Doesn't: A Regime Map and Its Mechanism

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhang, Haoxiang, Xu, Qixin, Li, Zhuofeng, Zhang, Lei, Jiang, Pengcheng, Zhang, Yu, McAuley, Julian
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913174986424320
author Zhang, Haoxiang
Xu, Qixin
Li, Zhuofeng
Zhang, Lei
Jiang, Pengcheng
Zhang, Yu
McAuley, Julian
author_facet Zhang, Haoxiang
Xu, Qixin
Li, Zhuofeng
Zhang, Lei
Jiang, Pengcheng
Zhang, Yu
McAuley, Julian
contents Long-horizon search agents accumulate large amounts of retrieved content across many tool calls, making context-budget efficiency increasingly important. A minimal intervention is to mask stale observations from the context as the trajectory progresses, but it remains unclear when this form of context management helps and why. We study observation masking through a systematic sweep over various agent backbones (4B to 284B parameters) and three retrievers on offline and live-web agentic search benchmarks. We find that the accuracy gain from masking follows an asymmetric inverted-U shape when plotted against the model's accuracy without context management: a plateau under weak retrievers, a peak when a strong retriever meets a mid-capacity model, and a sharp collapse when the model is saturated. This pattern reflects the interaction between retriever recall and the model's implicit filtering capacity, rather than either factor in isolation. Mechanistically, masking implements a token-for-turn trade-off: it removes observations the model has largely stopped attending to and pages the agent rarely re-opens. The added turns help when they convert failures into successes, but they fail when masking removes evidence the model would otherwise have used. We therefore reframe context management as a regime-dependent intervention and provide a holistic perspective for analyzing context use in agentic deep search. We release our scaffold and trajectories here (https://github.com/i-DeepSearch/observation-masking) to support future research.
format Preprint
id arxiv_https___arxiv_org_abs_2606_00408
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Masking Stale Observations Helps Search Agents -- Until It Doesn't: A Regime Map and Its Mechanism
Zhang, Haoxiang
Xu, Qixin
Li, Zhuofeng
Zhang, Lei
Jiang, Pengcheng
Zhang, Yu
McAuley, Julian
Computation and Language
Artificial Intelligence
Information Retrieval
Long-horizon search agents accumulate large amounts of retrieved content across many tool calls, making context-budget efficiency increasingly important. A minimal intervention is to mask stale observations from the context as the trajectory progresses, but it remains unclear when this form of context management helps and why. We study observation masking through a systematic sweep over various agent backbones (4B to 284B parameters) and three retrievers on offline and live-web agentic search benchmarks. We find that the accuracy gain from masking follows an asymmetric inverted-U shape when plotted against the model's accuracy without context management: a plateau under weak retrievers, a peak when a strong retriever meets a mid-capacity model, and a sharp collapse when the model is saturated. This pattern reflects the interaction between retriever recall and the model's implicit filtering capacity, rather than either factor in isolation. Mechanistically, masking implements a token-for-turn trade-off: it removes observations the model has largely stopped attending to and pages the agent rarely re-opens. The added turns help when they convert failures into successes, but they fail when masking removes evidence the model would otherwise have used. We therefore reframe context management as a regime-dependent intervention and provide a holistic perspective for analyzing context use in agentic deep search. We release our scaffold and trajectories here (https://github.com/i-DeepSearch/observation-masking) to support future research.
title Masking Stale Observations Helps Search Agents -- Until It Doesn't: A Regime Map and Its Mechanism
topic Computation and Language
Artificial Intelligence
Information Retrieval
url https://arxiv.org/abs/2606.00408