Line of inquiry
Inquiring lines›How do language models develop gen…›How do inference-time strategies a…›this line of inquiry
What causes foundation models to develop task heuristics instead of understanding?
A broader line of inquiry — a family of 25 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 25
Specific inquiring lines the field asks around this — ordered from the most general framing down to the most specific angle.
- Why do foundation models develop heuristics instead of world models?
- Why do foundation models develop task-specific heuristics instead of general world models?
- How do foundation models develop task-specific heuristics instead of world models?
- Can explicit constraint statements override the dominance of surface heuristics?
- Do foundation models develop task-specific shortcuts instead of building stable world models?
- Can surface heuristics override implicit constraints in domain-specific reasoning?
- What distinguishes task-specific heuristics from genuine world models?
- Can correct outputs mask reliance on surface heuristics rather than deep understanding?
- Why do foundation models rely on surface patterns instead of causal structure?
- Do task-specific heuristics improve gradually or appear suddenly at scale?
- Why do task-specific heuristics fail at generalizing to sparse data regions?
- Why do models fail on logically equivalent tasks with different data distributions?
- Does the heuristic dominance ratio vary predictably across model architectures?
- Why do foundation models develop task-specific heuristics instead of causal understanding?
- What inductive bias would force models to learn Newtonian mechanics instead of shortcuts?
- Do task-specific heuristics emerge because they compress well enough?
- Do models need world models to reliably steer hypothesis discovery?
- Does the pretrained prior actually constrain what internalized search can discover?
- How do surface statistical regularities enable correct outputs while degrading robustness?
- Can benchmarks designed for shortcut learning detect heuristic override failures?
- Can models adapt and combine search strategies beyond their training algorithm?
- How do unstated constraints become invisible to training data distributions?
- How do unstated feasibility constraints affect model decision-making?
- Why do good algorithms become rarer as the search space grows more generic?
- What distinguishes intrinsic search from extrinsic search method approaches?