ARTFEED — Contemporary Art Intelligence

DeltaPrompts: Escaping the Zero-Delta Trap in Multimodal Distillation

ai-technology · 2026-05-18

A new arXiv paper (2605.15532) reveals that up to 69% of prompts in standard chart/document reasoning datasets are zero-delta, meaning teacher and student VLMs produce identical answer distributions, providing minimal learning signal. The authors propose selecting prompts based on answer divergence to expose functional capability gaps.

Key facts

  • arXiv paper 2605.15532
  • Up to 69% of prompts in standard chart/document reasoning datasets are zero-delta
  • Zero-delta prompts cause student improvement to saturate regardless of data scale
  • Proposes selecting prompts based on answer divergence (Δ)
  • Non-zero divergence is critical for effective distillation

Entities

Institutions

  • arXiv

Sources