Test-Time Augmentation for LLMs: When Input Diversity Beats Output Diversity at Matched Compute

Name: Test-Time Augmentation for LLMs: When Input Diversity Beats Output Diversity at Matched Compute
Uploaded: 2026-08-11T05:59:37.400970+00:00
Duration: 1054 s
Description: The episode discusses a paper by AWS researchers comparing input diversity (rephrasing questions) to output diversity (multiple reasoning paths) for LLM inference. They find semantic rephrasing delivers 1.8x more accuracy per dollar than self-consistency, with gains on five of six tasks, but recommend it mainly for mid-tier models.

The episode discusses a paper by AWS researchers comparing input diversity (rephrasing questions) to output diversity (multiple reasoning paths) for LLM inference. They find semantic rephrasing delivers 1.8x more accuracy per dollar than self-consistency, with gains on five of six tasks, but recommend it mainly for mid-tier models.

Open the full episode