Line of inquiry
Inquiring lines›How do training choices shape mode…›How can retrieval systems better h…›this line of inquiry
How should reasoning time be allocated across search iterations?
A broader line of inquiry — a family of 32 specific questions the research asks around this. Follow one into its inquiring-line page, or move sideways to a related line below.
Questions in this line of inquiry 32
Specific inquiring lines the field asks around this — ordered from the most general framing down to the most specific angle.
- What is the optimal balance between search rounds and reasoning depth per round?
- Does unrestricted reasoning per search step degrade iterative quality over time?
- Can adaptive per-step decisions outperform uniform retrieval policies across different reasoning tasks?
- How should iterative research tasks limit context per reasoning turn?
- Do single-step retrieval systems with sophisticated synthesis qualify as deep research?
- How does search budget affect answer quality at test time?
- Why do per-turn thinking budgets matter alongside iterative retrieval depth?
- What limits exist on retrieval budget during inference?
- Why do per-turn reasoning caps improve iterative search quality?
- Why do deep research agents outperform retrieval augmented generation systems?
- Can step-level rewards improve training of agentic retrieval systems?
- What makes search budget matter for research task performance?
- How does overthinking in early turns degrade later retrieval rounds?
- How can per-step decisions about knowledge retrieval improve reasoning over uniform policies?
- What makes proactive tool retrieval better than single-round semantic matching?
- When should a system choose extended thinking versus quick responses?
- Can tree search improve question generation the way it improves reasoning?
- Does brute force experimentation substitute for research intuition and taste?
- What makes web retrieval more effective than static knowledge bases?
- Why do longer queries benefit less from clarification questions?
- How do search tasks differ from derivation tasks in reasoning efficiency?
- How does active learning reduce queries needed for user preference inference?
- Does RL pruning of documents differ fundamentally from rationale-driven evidence selection?
- Why does explicit reasoning degrade passage reranking performance?
- How does semantic search over research papers guide autonomous architecture proposals?
- What role does vague intent play in realistic search evaluation?
- Can attribute decomposition improve other interactive reasoning tasks beyond clinical questioning?
- Does high knowledge density in text reduce user motivation to read more?
- Why does GraphRAG prioritize corpus completeness while LogicRAG prioritizes query adaptivity?
- How do AIDE2's held-out gains compare to matched-budget test-time search baselines?
- What makes intent taxonomies unmanageable at hundreds of intents?
- How do real search queries reveal what counts as a deep research question?