Line of inquiry
Inquiring lines›How do training data and architect…›What does iterative improvement ac…›this line of inquiry
Does evolutionary optimization at test time improve model performance effectively?
A broader line of inquiry — a family of 45 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 45
Specific inquiring lines the field asks around this — ordered from the most general framing down to the most specific angle.
- How does the generation-verification gap shape evolutionary program search?
- Can evolved algorithms transfer learning strategies across different datasets and tasks?
- Can the same problem be solved by multiple evolutionary search strategies?
- How does test-time search budget compare to evolution gains under matched conditions?
- Does evolutionary inference transcend the parallel versus sequential test-time compute tradeoff?
- Can evolutionary search at inference time scale beyond natural language planning?
- Why do isolated evolutionary branches fail to propagate useful discoveries?
- Does co-evolution empirically outperform single-entity self-improvement in standard evaluations?
- How do evolutionary archives improve on single self-modification trajectories?
- Do evolutionary archives let agents improve themselves without formal proof?
- How do evolutionary archives enable diverse exploration in self-improving systems?
- Do evolutionary discovery systems like FunSearch count as bounded or open-ended improvement?
- Can harness evolution be redirected from memorization toward strategy distillation?
- Can harness evolution gains be distinguished from test-time search improvements on matched budgets?
- How much of harness-evolution gain comes from matched test-time search budgets?
- Why do evolutionary archives lead to more transferable discoveries than single-trajectory optimization?
- Does learned evaluator co-evolution solve the problem of verifying hard-to-benchmark tasks?
- What makes evolving the benchmark different from evolving the optimizer itself?
- Does population-based evolution transcend the parallel versus sequential compute tradeoff?
- Can evolutionary approaches avoid the overthinking failure mode of iterative refinement?
- Can LLM-based crossover and mutation work in unstructured natural language spaces?
- How does fitness-proportional selection guide LLM recombination in unstructured solution spaces?
- How does component-level self-evolution prevent information loss in multi-agent trajectories?
- Does AIDE2 archive rejected variants the way evolutionary approaches do for future reuse?
- Does co-evolution between peer agents reduce reliance on human design?
- Can accelerated sampling techniques from image generation speed up evolutionary search?
- How can diversity be preserved in evolving hypothesis populations?
- Can moving or evolving objectives prevent misalignment in discovery agents?
- Can co-evolved critics truly circumvent static evaluator limitations in self-improvement?
- What feedback signals matter most during harness evolution search?
- How does compiling natural language goals into executable code enable objective evolution?
- Why do evolutionary algorithms collapse to single solutions under selection pressure?
- Should we train the evolver or the executor when building self-improving agents?
- How does open-ended evolver reasoning identify patterns across heterogeneous user trajectories?
- Does this language evolution pattern occur outside the SaveVeyru scenario?
- Can evolutionary search unlock problems that best-of-n selection cannot solve?
- How does latent space diffusion enable evolutionary search in high dimensions?
- How does controlled utility evolution prevent the evaluator from becoming a new bottleneck?
- Is agentic efficiency analogous to convergent evolution in biology?
- Which parent-selection strategies improve hypothesis quality most?
- How does HGM's clade-based expansion differ from Darwin Gödel Machine's scoring approach?
- Why does island model genetic evolution maintain diversity better than single populations?
- Why does population-based search outperform both parallel and sequential test-time scaling?
- What interpretability challenges arise when algorithms are discovered rather than designed?
- How much of DarwinX's gain comes from maintaining an archive versus single-lineage search?