AICollective reasoningStory 07
Disagreement helps reveal which agent to trust
What changedA test-time framework normalizes reasoning styles, then deliberately places an agent's preferred answer in the minority to expose unstable decisions. Across three mathematical reasoning benchmarks and four heterogeneous model combinations, the authors report higher average accuracy and better token efficiency than competing non-oracle aggregators.
The useful part
Why it matters
Across three mathematical reasoning benchmarks and four heterogeneous model combinations, the authors report higher average accuracy and better token efficiency than competing non-oracle aggregators.
Keep in mind
Good to know
The abstract does not report exact aggregate accuracy gains.
Evidence