English static mirror for SEO/GEO · AI-assisted translation · Read Chinese original

Laser: Governing Long-Horizon Agentic Search via Structured Protocol and Context Register

Forum topic · 小凯 · 2026-07-05

Summary

Laser is a framework for stabilizing and scaling LLM-based agentic search, presented in an arXiv paper (arXiv:2512.20458) by Shuting Wang, Qiaolin Xia, and Zhicheng Dou and colleagues. Existing agentic search systems rely on unstructured natural-language reasoning and accumulate raw intermediate traces in context, causing unstable reasoning trajectories, context overflow, and degraded multi-hop QA performance. Laser addresses this with two components: (1) a symbolic action protocol that organizes agent behaviors into three spaces—planning, task-solving, and retrospection—with explicit semantics and deterministic execution formats for reliable parsing and traceable intermediate decisions; and (2) a compact context register that stores only essential reasoning states, enabling long-horizon reasoning without uncontrolled context expansion. Experiments on Qwen2.5/3-series models across challenging multi-hop QA datasets show Laser consistently outperforms existing agentic search baselines under both prompting-only and fine-tuning settings, offering a principled foundation for robust, scalable agentic search.

Laser: Governing Long-Horizon Agentic Search via Structured Protocol and Context Register

Paper: https://arxiv.org/abs/2512.20458 — published 2025-12-23, in the Agentic Search category.

Authors: Shuting Wang, Qiaolin Xia, Vich Wang, Herberttli, Bobsimons, Zhicheng Dou

Overview

Agentic search systems built on Large Language Models (LLMs) and Large Reasoning Models (LRMs) interleave multi-step reasoning with external tool use. However, existing frameworks largely rely on unstructured natural-language reasoning and accumulate raw intermediate traces in the context, which leads to unstable reasoning trajectories, context overflow, and degraded performance on complex multi-hop queries.

Laser is a general framework for stabilizing and scaling agentic search, built on two core mechanisms:

1. Symbolic action protocol — Agent behaviors are organized into three spaces: planning, task-solving, and retrospection. Each action has explicit semantics and a deterministic execution format, enabling structured reasoning and reliable action parsing. Intermediate decisions become interpretable and traceable, supporting explicit retrospection and fine-grained control over reasoning trajectories.

2. Compact context register — In coordination with parsable actions, Laser maintains a register that stores only essential states of the reasoning process, allowing the agent to reason over long horizons without uncontrolled context expansion.

Experimental Results

Experiments on Qwen2.5/3-series models across challenging multi-hop QA datasets show that Laser consistently outperforms existing agentic search baselines under both prompting-only and fine-tuning settings, demonstrating a principled and effective foundation for robust, scalable agentic search.

Original Abstract

> Recent advances in Large Language Models (LLMs) and Large Reasoning Models (LRMs) have enabled agentic search systems that interleave multi-step reasoning with external tool use. However, existing frameworks largely rely on unstructured natural-language reasoning and accumulate raw intermediate traces in the context, which often leads to unstable reasoning trajectories, context overflow, and degraded performance on complex multi-hop queries. In this study, we introduce Laser, a general framework for stabilizing and scaling agentic search. Laser defines a symbolic action protocol that organizes agent behaviors into three spaces: planning, task-solving, and retrospection. Each action is specified with explicit semantics and a deterministic execution format, enabling structured and logical reasoning processes and reliable action parsing. This design makes intermediate decisions interpretable and traceable, enhancing explicit retrospection and fine-grained control over reasoning trajectories. In coordination with parsable actions, Laser further maintains a compact context register that stores only essential states of the reasoning process, allowing the agent to reason over long horizons without uncontrolled context expansion. Experiments on Qwen2.5/3-series models across challenging multi-hop QA datasets show that Laser consistently outperforms existing agentic search baselines under both prompting-only and fine-tuning settings, demonstrating that Laser provides a principled and effective foundation for robust, scalable agentic search.

Related Reading

Tags

#agentic-search#llm#reasoning#multi-hop-qa#context-management#rag#qwen#arxiv

This page is an English static mirror generated for search and AI citation. It may be a full translation or structured summary of the Chinese original. Canonical interactive discussion lives on the Chinese page: https://zhichai.net/topic/178208473