Line of inquiry
Inquiring lines›How do training data and architect…›What training signals and data sel…›this line of inquiry
When and why does aggregating multiple reasoning samples improve performance?
A broader line of inquiry — a family of 27 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 27
Specific inquiring lines the field asks around this — ordered from the most general framing down to the most specific angle.
- Why does majority voting reward work better than other test-time aggregation methods?
- How does training-time voting differ from inference-time majority voting over samples?
- How does majority voting fail when reasoning samples lack genuine diversity?
- Does majority voting reliably signal correctness without risking reward hacking?
- When does multi-agent voting help versus hurt performance on tasks?
- Can test-time voting improve reasoning beyond the base model's original capabilities?
- Does majority voting prevent confident but incorrect answers from being reinforced?
- How does training-time consensus differ from inference-time majority voting over samples?
- Can voting work at every level of task decomposition, not just whole problems?
- What compute costs does majority-vote consensus sampling add versus supervised training?
- What makes consensus games work without retraining the base model?
- What happens when majority voting converges to a single answer?
- What mechanisms let generative models escape collapse through majority voting?
- Can a policy game vote-based rewards through distinguishability unrelated to quality?
- How do correlated errors across agents threaten voting-based error correction systems?
- What intermediate information does majority voting discard from reasoning chains?
- Why do majority-vote rewards amplify errors below an accuracy threshold?
- Why do self-consistency methods fail where pretraining bias is strongest?
- Can crowdsourced voting reliably identify correct answers on graduate-level factual questions?
- How does majority vote consensus handle cases where the consensus is confidently wrong?
- Can aggregation cancel out independent errors but not shared training-data biases?
- Why does low temperature sampling extract consensus from diverse training data?
- What signals detect when consensus training is silently degrading performance?
- Why do preference-optimized AI systems converge on consensus hypotheses instead of exploring alternatives?
- Why does training on agreement signals between samples differ from selecting among them?
- Can calibrated confidence reduce misleading consensus in group deliberation?
- Which prompt properties determine whether variance helps under majority voting?