AI RESEARCH
Compute Allocation for Reasoning-Intensive Retrieval Agents
arXiv CS.AI
•
ArXi:2603.14635v1 Announce Type: cross As agents operate over long horizons, their memory s grow continuously, making retrieval critical to accessing relevant information. Many agent queries require reasoning-intensive retrieval, where the connection between query and relevant documents is implicit and requires inference to bridge. LLM-augmented pipelines address this through query expansion and candidate re-ranking, but