Line of inquiry
Inquiring lines›How do agents behave and coordinat…›How do multi-agent systems compare…›this line of inquiry
What enables multi-agent systems to outperform single-model approaches?
A broader line of inquiry — a family of 67 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 67
Specific inquiring lines the field asks around this — ordered from the most general framing down to the most specific angle.
- When does multi-agent scaling actually outperform static ensembles?
- Can we design efficient agents by targeting constraints directly?
- How do multi-agent systems improve on single frontier models?
- How do perception and execution gaps limit current AI agent performance?
- What makes planning, tool use, and reasoning into jointly optimizable subsystems?
- Does internal task decomposition eliminate overhead from multi-agent coordination?
- Which research tasks are better suited for multi-agent versus single-agent approaches?
- What makes composable abstractions emerge under performance pressure in agent systems?
- At what task difficulty does multi-agent decomposition become worth the coordination cost?
- Does parallel task structure determine optimal multi-agent architecture?
- How do planning and grounding have opposing optimization requirements in agents?
- Can multi-agent teams solve problems better than single models thinking longer?
- Can cognitive diversity compensate for lack of expertise in agent teams?
- How should we measure context efficiency and verification cost in agents?
- What accounts for performance drops in multi-turn agent interactions?
- Can cognitive diversity overcome expertise gaps in agent teams?
- How do multi-agent routers balance flexibility against interpretability in design?
- How do static team decomposition and dynamic agent selection compare in efficiency?
- What four decisions matter most in multi-agent system routing?
- Which layer of agent systems creates the largest capability gains in practice?
- How does role allocation in multi-agent systems depend on model differentiation?
- Can code-based reasoning replace natural language deliberation in agentic systems?
- Can the scaling law for discovery extend beyond architectures to agentic systems?
- When does forcing agent reasoning into code become a leaky abstraction?
- How should proportionality constraints be implemented in agentic systems?
- How does role specialization preserve reasoning diversity in multi-agent teams?
- How do capability vectors enable discovery in multi-agent systems?
- Can agents balance goal-driven proactivity with user preference alignment?
- Do information gathering and task execution require different incentive structures?
- How does deterministic feature engineering increase information for computationally bounded agents?
- Can task decomposition into microagents with voting scale to million-step problems?
- Should optimal context budgets scale with agent competence or task complexity?
- Can construction-time routing and runtime agent pruning be combined effectively?
- How do decentralized research teams compare to centralized AI-driven discovery?
- Why does capability discovery become the bottleneck in large agent systems?
- How do cognitive stimulation and process losses interact in group AI systems?
- How will the agent economy reshape compute infrastructure design?
- When should you optimize agent behavior versus tool performance separately?
- Can heterogeneous AI agents integrate through shared API and MCP interfaces?
- How does multi-agent reasoning scale compared to single-model approaches?
- How would a bi-level agent restructure objective functions during discovery?
- What makes capability vectors a better coordination substrate than topic-based routing?
- What structural features drive instrumental convergence across different agent goals?
- Does cognitive diversity in teams only pay off when agents actively explore it?
- Can we decompose agent efficiency into measurable independent components?
- Why does literature review benefit most from multi-agent orchestration approaches?
- Can language agents be represented as optimizable computational graphs?
- Why do planning and grounding have opposing optimization requirements in agents?
- Can backward planning reduce search difficulty when multiple goal state paths exist?
- How do language agents become optimizable computational graphs automatically?
- What role does exploration-exploitation balance play in abstraction formation?
- Why has agent research prioritized policy over world model development?
- Can multi-agent reasoning systems scale beyond current architectures?
- What scaling laws govern autonomous architecture discovery in AI systems?
- When should agents stop recursing to optimize success versus cost?
- Can weaker planners match stronger models if behavior is reorganized?
- Does the planning-grounding factoring principle apply to other agent tasks?
- What makes a service visible to autonomous agent systems?
- Can objective search escape the limitations of fixed-objective central planning?
- Is agentic efficiency analogous to convergent evolution in biology?
- Why does decentralization work better than central planning for open-ended research?
- How should topology routing adapt to different task types?
- How should experiment budgets be allocated across parallel hypothesis-testing teams?
- How do sharded HNSW indices preserve capability distinctions at scale?
- What ecosystem conditions make agent attention markets viable?
- What capability threshold do agents need to self-organize effectively?
- What structural constraints produce recursion costs in agentic systems?