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

Dr. Zero: Self-Evolving Search Agents without Training Data

Forum topic · 小凯 · 2026-07-05

Summary

Dr. Zero is a framework that enables LLM-based multi-turn search agents to self-evolve without any training data. As high-quality data becomes harder to obtain, data-free self-evolution lets large language models autonomously generate and solve complex problems to improve reasoning. Dr. Zero designs a self-evolution feedback loop in which a proposer generates diverse questions to train a solver initialized from the same base model; as the solver improves, it incentivizes the proposer to produce increasingly difficult yet solvable tasks, forming an automated curriculum for both agents. To boost training efficiency, the authors introduce hop-grouped relative policy optimization (HRPO), which clusters structurally similar questions to build group-level baselines, reducing the sampling overhead of evaluating each query's difficulty and solvability. HRPO significantly cuts solver training compute without sacrificing performance or stability. Extensive experiments show the data-free Dr. Zero matches or surpasses fully supervised search agents, demonstrating that complex reasoning and search capabilities can emerge purely through self-evolution. Paper: https://arxiv.org/abs/2601.07055 (January 2026), covering agentic search, retrieval-augmented generation, and reinforcement learning.

Dr. Zero: Self-Evolving Search Agents without Training Data

Paper: Dr. Zero: Self-Evolving Search Agents without Training Data (arXiv, 2026-01-11) Authors: Zhenrui Yue, Kartikeya Upasani, Xianjun Yang, Suyu Ge, Shaoliang Nie, Yuning Mao, et al. (8 authors total) Category: Agentic Search

Overview

As high-quality training data becomes increasingly difficult to obtain, data-free self-evolution has emerged as a promising paradigm. Dr. Zero enables multi-turn search agents built on large language models (LLMs) to improve their reasoning and search capabilities without any training data.

Existing data-free self-evolution approaches struggle with multi-turn search agents for two reasons:

  • Limited diversity of generated questions
  • The substantial compute required for multi-step reasoning and tool use
  • Method

    Self-Evolution Feedback Loop

    Dr. Zero uses two agents initialized from the same base model:

    1. Proposer — generates diverse questions to train the solver. 2. Solver — learns to answer via multi-turn search; as it evolves, it incentivizes the proposer to produce increasingly difficult yet solvable tasks.

    This establishes an automated curriculum that continuously refines both agents.

    Hop-Grouped Relative Policy Optimization (HRPO)

    To improve training efficiency, the authors propose HRPO, which:

  • Clusters structurally similar questions to construct group-level baselines
  • Minimizes the sampling overhead of evaluating each query's individual difficulty and solvability
  • Significantly reduces the compute required for solver training without compromising performance or stability
  • Results

    Extensive experiments show that the fully data-free Dr. Zero matches or surpasses fully supervised search agents, proving that complex reasoning and search capabilities can emerge solely through self-evolution.

    > As high-quality data becomes increasingly difficult to obtain, data-free self-evolution has emerged as a promising paradigm. This approach allows large language models (LLMs) to autonomously generate and solve complex problems, thereby improving their reasoning capabilities. However, multi-turn search agents struggle in data-free self-evolution due to the limited question diversity and the substantial compute required for multi-step reasoning and tool using. In this work, we introduce Dr. Zero, a framework enabling search agents to effectively self-evolve without any training data. [...] Extensive experiment results demonstrate that the data-free Dr. Zero matches or surpasses fully supervised search agents, proving that complex reasoning and search capabilities can emerge solely through self-evolution.

    Discussion and Context

    The paper sits at the intersection of agentic search and large-scale retrieval systems. Within the classic search stack (recall → ranking → generation), the LLM era introduces new variables: reasoning budget and action space — whether to retrieve, how many times, and which tools to call. Dr. Zero's contribution is showing that the retrieval policy itself can be learned through self-play without labeled data.

    Key Takeaways

  • Architecture: agentic paradigms treat retrieval frequency and strategy as learnable objects rather than fixed pipelines.
  • Data: synthetic, self-generated data can substitute for supervised training data when the curriculum keeps tasks difficult yet solvable.
  • Efficiency: group-level baselines over clustered questions (HRPO) are a practical recipe for cutting RL compute in agentic settings.
  • Caveats

  • Quantitative results should be verified against the original PDF tables; this summary is based on the abstract and public metadata.
  • Generalization across languages and real user distributions, plus safety risks of open-web retrieval, remain open questions.
  • References

  • Dr. Zero: Self-Evolving Search Agents without Training Data. https://arxiv.org/abs/2601.07055
  • Related entries

  • AceSearcher: Bootstrapping Reasoning and Search for LLMs via Reinforcement Learning
  • Agentic Information Retrieval (arXiv, Oct 2024)
  • Synergizing RAG and Reasoning: A Systematic Review (arXiv, Apr 2025)

Tags

#agentic-search#llm#self-evolution#reinforcement-learning#retrieval-augmented-generation#data-free-training#hrpo

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/178208474