HybridResearcher: Bridging the Structured-Unstructured Boundary for Deep Research Agents

Hugging Face Daily Papers Papers
deep-research-agents cross-source-reasoning reinforcement-learning synthetic-data text-to-sql small-language-models

Summary

论文提出 HybridResearcher，一个 4B 小型深度研究智能体，通过合成跨源轨迹与强化学习在结构化与非结构化数据之间进行推理，在 HybridDeepResearch 基准上超越更大的 Qwen3.5-397B-A17B，同时不遗忘单源能力。

While LLM agents are increasingly capable of gathering and synthesizing evidence from a single homogeneous source, such as structured tables or unstructured data, real-world analytical questions frequently span heterogeneous sources. For example, answering one enterprise question may require both a result computed over an internal database and evidence found on the web. Recent evaluations on benchmarks like HybridDeepResearch show that even frontier agents struggle when evidence must cross this boundary. While these benchmarks have exposed the limitation, methodologies to explicitly teach cross-source reasoning remain absent. To address this, we introduce HybridResearcher, a 4B agent trained to bridge evidence across the structured and unstructured boundary. To build it, we synthesize cross-source bridge trajectories by starting with validated single-source tasks-such as existing text-to-SQL or web search queries-and fabricating the missing source around a bridge entity drawn from the seed's own output. We retain a sample only when strict construction gates confirm that neither side alone suffices, ensuring the data requires true cross-source interdependence. Finally, we train on this curated dataset using supervised fine-tuning followed by reinforcement learning. Evaluated on the existing HybridDeepResearch benchmark, HybridResearcher improves over our base model, Qwen3.5-4B, by 30.0 points pass@8 and surpasses the far larger Qwen3.5-397B-A17B by 8.3 points, while also improving over its base model on established, held-out single-source benchmarks (BrowseComp and BrowseComp-Plus), demonstrating that this complex cross-source reasoning can be taught directly to small models without triggering catastrophic forgetting.
Original Article
View Cached Full Text

Cached at: 10/21/26, 07:02 AM

Paper page - HybridResearcher: Bridging the Structured-Unstructured Boundary for Deep Research Agents

Source: https://huggingface.co/papers/2610.24899 Authors:

,

,

,

,

,

,

,

,

,

Abstract

HybridResearcher is a small agent trained via synthetic cross-source trajectories and reinforcement learning to reason across structured and unstructured data, outperforming much larger models without forgetting single-source skills.

WhileLLM agentsare increasingly capable of gathering and synthesizing evidence from a single homogeneous source, such as structured tables or unstructured data, real-world analytical questions frequently span heterogeneous sources. For example, answering one enterprise question may require both a result computed over an internal database and evidence found on the web. Recent evaluations on benchmarks like HybridDeepResearch show that even frontier agents struggle when evidence must cross this boundary. While these benchmarks have exposed the limitation, methodologies to explicitly teachcross-source reasoningremain absent. To address this, we introduce HybridResearcher, a 4B agent trained to bridge evidence across the structured and unstructured boundary. To build it, we synthesize cross-source bridge trajectories by starting with validated single-source tasks-such as existingtext-to-SQLorweb searchqueries-and fabricating the missing source around a bridge entity drawn from the seed’s own output. We retain a sample only when strict construction gates confirm that neither side alone suffices, ensuring the data requires true cross-source interdependence. Finally, we train on this curated dataset usingsupervised fine-tuningfollowed byreinforcement learning. Evaluated on the existing HybridDeepResearch benchmark, HybridResearcher improves over our base model, Qwen3.5-4B, by 30.0 points pass@8 and surpasses the far larger Qwen3.5-397B-A17B by 8.3 points, while also improving over its base model on established, held-out single-source benchmarks (BrowseComp and BrowseComp-Plus), demonstrating that this complexcross-source reasoningcan be taught directly to small models without triggeringcatastrophic forgetting.

View arXiv pageView PDFAdd to collection

Get this paper in your agent:

hf papers read 2610\.24899

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2610.24899 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2610.24899 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2610.24899 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

PrimeSeeker: Capability-Oriented Supervision for Deep Search Agents

arXiv cs.CL

PrimeSeeker 提出一种面向能力的监督框架，通过「潜在锚点推理」将深度搜索分解为锚点解析与关系传递的耦合操作，据此构建网络证据结构、联合生成问题与参考证据骨架，训练出 30B 搜索智能体，在五个深度搜索基准上以更少工具调用取得强性能。

Hybrid Open-Ended Tri-Evolution Makes Better Deep Researcher

arXiv cs.AI

This paper proposes the Hybrid Open-Ended Tri-Evolution (HOTE) framework, which uses hybrid-mode reinforcement learning to evolve a proposer, solver, and judge collaboratively for deep research tasks, achieving state-of-the-art results with an 8B model surpassing larger static models.