跳到正文
HuggingFace Daily Papers(社區熱門論文)·· 2026-10-17AI 評分46

HybridResearcher:打通深度研究智能體的結構化與非結構化數據邊界

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

AI 導讀

HybridResearcher 是一款以合成跨來源軌跡和強化學習訓練的 4B 智能體,能跨結構化與非結構化數據推理。在 HybridDeepResearch 基準上,其 pass@8 比基礎模型 Qwen3.5-4B 高 30.0 分,並超越規模大得多的 Qwen3.5-397B-A17B 8.3 分;在 BrowseComp 和 BrowseComp-Plus 上亦優於基礎模型。

正文

Published on Oct 17

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.

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.

View arXiv page View PDF Add 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 paper 0

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 paper 0

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 paper 0

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 paper 0

No Collection including this paper

Add this paper to a collection to link it from this page.

來源:HuggingFace Daily Papers(社區熱門論文) · huggingface.co