On the Limits of Scale: Inverse Scaling in Vision-Language Models
Scaling laws suggest that increasing model size, training data, and compute generally yields monotonic performance improvements. While this often holds at the aggregate level, we present systematic evidence that vision-language models (VLMs) exhibit inverse scaling at the subtask level: specific…
By Rohit Saxena, Alessandro Suglia, Marek Rei, Pasquale MinerviniarXiv - cs.CV (CV)
Read the full story
Originally published by arXiv - cs.CV (CV) on Oct 21, 2026.
Comments
Sign in to join the discussion.