AI RESEARCH
Diversity Matters: Revisiting Test-Time Compute in Vision-Language Models
arXiv CS.LG
•
ArXi:2605.30713v1 Announce Type: new Test-time compute (TTC) strategies have emerged as a lightweight approach to boost reasoning in large language models (LLMs). However, their application and benefits for vision-language models (VLMs) remain underexplored. We present a systematic study of TTC across seven VLMs and six benchmarks, specifically analyzing feature-based scoring and majority voting methods. We find that feature heuristics fail and voting yields only modest gains in single-model settings.