What Language Adds, and at What Cost: Four Controlled Probes of Pathology Vision-Language Models

Key points

• We grade published evidence by how it was produced and test four claims with controlled probes that reuse released checkpoints and public data, varying one input at a time.1
• Report text adds prognostic signal: adding a frozen embedding of the stage-redacted report to a slide embedding raises the pooled C-index by 0.019 for progression-free interval and by 0.016 for overall survival across 1{,}633 TCGA patients, and restoring the stage statement adds nothing.1
• Prompts add no measurable accuracy at inference: with features and pooling fixed, zero-shot classification is statistically indistinguishable from a supervised linear probe for lung and renal cancer subtyping.1
• Self-reported gains from agentic navigation and few-shot language priors cannot yet be attributed to language.1

Sources (1)

• [1]What Language Adds, and at What Cost: Four Controlled Probes of Pathology Vision-Language Models
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 19, 05:48 PM

  Vision--language models (VLMs) promise to make computational pathology queryable and able to draft reports, but how much language adds beyond the vision backbone is unclear, because almost every system is evaluated by its developers on test sets of their own choosing.

  We grade published evidence by how it was produced and test four claims with controlled probes that reuse released checkpoints and public data, varying one input at a time.

Extractive summary: sentences quoted from the sources.