Line of inquiry
Inquiring lines›How do training data and architect…›How should model design balance ar…›this line of inquiry
When does routing outperform scaling as a performance solution?
A broader line of inquiry — a family of 42 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 42
Specific inquiring lines the field asks around this — ordered from the most general framing down to the most specific angle.
- Can routing ever beat model improvement as a performance lever?
- Can multiple small models outperform a single large model with good routing?
- What makes routing a better investment than training larger models?
- Does model selection matter more than model improvement for query routing?
- Should model routing decisions account for prompt-tier dependencies?
- How does routing decide between models before generation happens?
- Can routing policies remain meaningful over behaviorally homogeneous model pools?
- Which components of LLM routers actually drive performance differences?
- Why might diverse smaller models with routing beat one giant model?
- How do routing and test-time compute scaling work together as optimization axes?
- How much worse are convergent routing methods than theoretical oracle routing?
- Can embedding-cluster routing outperform a single frontier model?
- Does curating smaller model pools improve router recall compared to scaling?
- Can routing enable heterogeneous SLM-first architectures at scale?
- Can routing systems prevent expert models from failing outside their specialty?
- What makes query complexity a better routing signal than response quality?
- Why do routers fail when only a few models have the right answer?
- How do pre-training and distillation enable minimal routing signals to work?
- Can clustering-based routing match neural network routers in cost and quality?
- Why does single-model routing beat ensemble and cascade approaches on latency?
- Can a router predict query complexity well but still fail as governance?
- Can model routing and compute allocation work together as independent optimizations?
- What makes routing a governance mechanism rather than just a predictor?
- Why do commercial LLM routers fail to beat lightweight baselines?
- How do routers decide when to escalate from small to large models?
- How do KNN and prompted routers differ in the accuracy-stability tradeoff?
- Can semantic routing couple similarity matching with resource constraints?
- Does the improved model actually return to the routing pool and shape future decisions?
- Can routing signals organize training data into a meaningful curriculum automatically?
- Does surface-form query rewriting allow attackers to steer model routing decisions?
- Does model capability still matter once coordination infrastructure is optimized?
- How do cost-aware cascades compare to single-turn routing in the component framework?
- What makes mixture-of-experts routing learn token-level specialization effectively?
- What learning signals best supervise router training across benchmark tasks?
- Why do KNN routers collapse when queries are paraphrased?
- How much safety burden shifts between provider filters and model alignment in rerouted requests?
- Why does Branch-Train-Merge fail without learned routing between experts?
- Who serves as the teacher model in the routing-guided distillation process?
- How much does evaluation methodology affect routing research claims?
- How does semantic clustering help decide which model handles each query?
- How much latency cost does query cascading actually add?
- Why does a second routing level sometimes break accuracy in disclosure hierarchies?