HybridResearcher: Bridging the Structured-Unstructured Boundary for Deep Research Agents
To address this, we introduce HybridResearcher, a 4B agent trained to bridge evidence across the structured and unstructured boundary.
ProofPaper ↗
Key points
- 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.
- Recent evaluations on benchmarks like HybridDeepResearch show that even frontier agents struggle when evidence must cross this 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.
- 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.
Sources (2)
- [1]HybridResearcher: Bridging the Structured-Unstructured Boundary for Deep Research AgentsarXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 17, 10:46 PM
To address this, we introduce HybridResearcher, a 4B agent trained to bridge evidence across the structured and unstructured boundary.
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.
- [2]HybridResearcher: Bridging the Structured-Unstructured Boundary for Deep Research AgentsHugging Face Daily Papers · Oct 17, 12:00 AM · same content
Extractive summary: sentences quoted from the sources.
Before this
- Oct 16, 2026MetaCurate: Evolving Curation Filters for High-Quality Agentic Coding Data
- Oct 16, 2026Welcome Beam: a 501B open model you can own, customize, and scale
- Oct 15, 2026AlphaBet: Reinforcement Learning for LLM Event Forecasting Agents
- Oct 15, 2026KnowGPO: Knowledge-Grouped Policy Optimization for Search Agents
- Oct 15, 2026Librarian Agents: Training LLMs to Curate Long-Context Workspaces that Anticipate Future Queries
- Oct 15, 2026Selection-Based Structured Reasoning: Toward Efficient Multimodal Search Agents