Google: PaLM 2 multilingual mixed-objective training system

What changed

Combined diverse multilingual data and mixed pretraining objectives with compute-aware model/data allocation, reporting a more inference-efficient language-model family and a larger-scale scaling study.

Scientific scope

Combined diverse multilingual data and mixed pretraining objectives with compute-aware model/data allocation, reporting a more inference-efficient language-model family and a larger-scale scaling study. The scaling allocation follows and tests Chinchilla rather than inventing its law; mixed objectives build on UL2. Scaling-study model sizes are not undisclosed PaLM 2 sizes. No medical-specialist variant is included.

Evidence

The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. The scaling allocation follows and tests Chinchilla rather than inventing its law; mixed objectives build on UL2. Scaling-study model sizes are not undisclosed PaLM 2 sizes. No medical-specialist variant is included.

Contributors

• Source-listed authors and technical contributors · SCIENCE: Source-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities.

• PaLM 2 Technical Report
• Google: PaLM 2 multilingual mixed-objective training system — official research record

Provisional WRS assessment, edition 2026-10-16.1. Independent expert calibration is pending. This grade does not certify correctness or assign ownership.

Full scientific record · Browse contribution grades · Request a correction