Dossier: Deep Research via Ledger-Driven Branching Search and Query Encoding Learning

Dossier is a deep research agent that replaces linear search trajectories with parallel branching search, tracking claims, contradictions, and information gaps in a persistent Research Ledger. It consistently outperforms ReAct-style agents on multi-hop research tasks by preventing early retrieval failures from compounding through the reasoning chain.

Abstract

Deep research requires synthesizing information across fragmented sources. Existing ReAct-style agents for such multi-hop retrieval typically rely on long, linear search trajectories, where early retrieval failures compound through the reasoning chain. We introduce Dossier, a deep research agent that replaces linear paths with locally parallel, branching search managed by a persistent Research Ledger. The Ledger explicitly tracks claims, contradictions, and information gaps, continuously updating as new evidence is synthesized. To improve the quality of each retrieval step, we introduce Evidence-Aligned Query Learning (EAQL), a training mechanism that fine-tunes query encoders to condition on the Research Ledger. This ensures that generated queries are contextually grounded in the agent’s evolving state rather than isolated prompts. Our evaluation demonstrates that Dossier’s branching architecture improves end-to-end accuracy on BrowseComp-Plus by 23 percentage points and HoVer by 29 percentage points across multiple LLM models