Line of inquiry
Inquiring lines›How should we train models for cap…›What systematic failures and vulne…›this line of inquiry
What determines success in training models on multiple tasks?
A broader line of inquiry — a family of 34 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 34
Specific inquiring lines the field asks around this — ordered from the most general framing down to the most specific angle.
- Can training on diverse related tasks be more efficient than task-specific training?
- Can a single model trained on two tasks predict untrained decision tasks?
- Does task superposition explain how models learn from multiple in-context trajectories?
- Why does full multi-task fine-tuning perform worse than sequential training?
- What performance trade-offs emerge when composing multiple independently trained model capabilities?
- How do neural networks decompose tasks into modular subnetworks that transfer?
- Why do larger models reduce interference between rare and common tasks?
- Can granular sub-task training for function calling improve both open and proprietary models?
- Does training on granular tasks beat training on the full function calling problem?
- Can models maintain multiple task interpretations simultaneously before committing to a single policy?
- Do different function-calling subtasks have different entropy profiles during training?
- How does task decomposition prevent bias from spreading across therapeutic AI pipelines?
- Does task ordering affect multi-task reinforcement learning outcomes?
- How should multi-objective post-training balance competing behavioral goals?
- When should model isolation be preferred over weight-averaging approaches?
- How much of the combinatorial task space must training data cover?
- Can sub-task handlers be swapped between neural and symbolic systems?
- What role does consensus merging play in dynamic task decomposition?
- Can we predict which tasks will decompose into modular subnetworks?
- What task structures benefit most from geometric parameter merging?
- How do complete multi-turn trajectories differ from isolated task examples?
- Can intentional data-mixture design replace model scaling for rare task learning?
- When and what should a model actually decide to delegate?
- How do ensemble methods apply within a single model?
- What happens when a single loss function conflates representation learning with decision-making?
- Can extracted skills transfer effectively across different domains and model architectures?
- How does joint backpropagation differ from training separate ensemble models?
- How do neural networks decompose complex tasks into modular subnetworks?
- Do interaction effects between research mechanisms depend on the task domain?
- How do transformers stitch together learned behaviors when adapting to new tasks?
- Why do models that excel at task success often fail at privacy compliance?
- How do gradients flowing through both branches simultaneously reshape each component's role?
- Can backward transfer measurements reliably predict optimal multi-task training order?
- How do orthogonal adapter vectors avoid interference at scale?