W29 Product Update: AI review false positives down 59.7%

AI reviews got more accurate: false positives dropped by 59.7%.

A false positive is a review that marks a correct translation as failing – fewer of them means fewer alerts to check by hand, and review scores that can be trusted. We measured the drop on our internal benchmark. The biggest change: when a review doesn't apply to a given string, it no longer counts against the score, so scores reflect only the reviews that genuinely apply.

The reviewer setup form now shows example instructions as a placeholder for each review type.

Also this week:

- Xcode String Catalogs in the GitHub App. The GitHub App now handles .xcstrings files – a String Catalog committed to a connected repository gets localized on every push, the same way the CLI and Async Localization API do.
- Source file path in request logs. Request logs from CLI runs now record which file each request came from, so a log entry traces back to the file that produced it.
- Key names guide translation in JSON and JSONC files. Translations use the real key names as context, so a key like checkout.error.cardDeclined informs the result. Applies across the CLI, the GitHub App, and the Async Localization API.

Full changelog: lingo.dev/changelog