CORE: Coarse-to-Fine Search with Context Management for Deep Research
cs.CL, cs.AI, cs.IR
Submitted: 2026-11-05
Updated: 2026-11-05
Comments: 24 pages
License: http://creativecommons.org/licenses/by/4.0/
The gist: Deep research agents answer complex questions through repeated search.
Terminology
Abstract
Deep research agents answer complex questions through repeated search. They rely on a harness that runs their tool calls and returns evidence to their context. This creates a trade-off: Search works best over long passages, which give enough context for reliable semantic matching. But the returned evidence should be short, because every observation stays in the agent's context for the rest of the trajectory. We introduce CORE, a training-free harness that combines coarse-to-fine search with context management. CORE retrieves passages with a dense retriever. It then reuses the token-level similarities from a late-interaction reranker to return only the most relevant snippet from each passage. Context management lets the agent expand the context around sources it has already retrieved and suppresses repeated snippets. In controlled closed-search comparisons with the Tongyi DeepResearch and OpenResearcher harnesses, CORE achieves state-of-the-art answer accuracy and evidence recall across three agent models and three benchmarks (BrowseComp-Plus, BrowseComp-Wikipedia and BrowseComp-Plus-CM), on corpora containing up to 553M documents.
Sources
- OpenSeeker: Democratizing Frontier Search Agents by Fully Open-Sourcing Training Data
- Rethinking Agentic Search with Pi-Serini: Is Lexical Retrieval Sufficient?
- Harness-1: Reinforcement Learning for Search Agents with State-Externalizing Harnesses
- WebSailor: Navigating Super-human Reasoning for Web Agent
- Search-o1: Agentic Search-Enhanced Large Reasoning Models
- OpenResearcher: A Fully Open Pipeline for Long-Horizon Deep Research Trajectory Synthesis
- WebExplorer: Explore and Evolve for Training Long-Horizon Web Agents
- Dr-DCI: Scaling Direct Corpus Interaction via Dynamic Workspace Expansion
- CMASS: A ClimbMix Agentic Search Suite for Evaluating Search Agents Across Heterogeneous Tasks
- Rerank Before You Reason: Analyzing Reranking Tradeoffs through Effective Token Cost in Deep Search Agents
- Deep Research: A Systematic Survey
- Tongyi DeepResearch Technical Report
- Search, Inspect, Fetch: Exploiting Structure-Aware Boolean Retrieval for Deep-Search Agents
- BrowseComp: A Simple Yet Challenging Benchmark for Browsing Agents
- ReSum: Unlocking Long-Horizon Search Intelligence via Context Summarization
- TokenPilot: Cache-Efficient Context Management for LLM Agents
- A Comprehensive Survey of Deep Research: Systems, Methodologies, and Applications
- Superintelligent Retrieval Agent: The Next Frontier of Agentic Retrieval
- AgentFold: Long-Horizon Web Agents with Proactive Context Management
- Masking Stale Observations Helps Search Agents -- Until It Doesn't: A Regime Map and Its Mechanism
Related papers
- ALARM: Audio-Language Alignment for Reasoning Models
- ThinkSum: Probabilistic reasoning over sets using large language models
- Coherence boosting: When your pretrained language model is not paying enough attention
- From Facts to Insights: A Persona-Driven Dual Memory Framework and Dataset for Role-Playing Agents
- Improving General Role-Playing Agents via Psychology-Grounded Reasoning and Role-Aware Policy Optimization
- The Art of Asking: Multilingual Prompt Optimization for Synthetic Data