Schick and colleagues: Toolformer self-supervised API learning

What changed

Generated and filtered API-call annotations by their effect on language-model prediction loss, then trained a model to select and use external tools.

Scientific scope

Generated and filtered API-call annotations by their effect on language-model prediction loss, then trained a model to select and use external tools. The original method requires a few demonstrations and evaluates specified APIs. Its paper notes limits on chained and interactive tool use; no general autonomy or correctness guarantee is implied.

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 original method requires a few demonstrations and evaluates specified APIs. Its paper notes limits on chained and interactive tool use; no general autonomy or correctness guarantee is implied.

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.

• Toolformer: Language Models Can Teach Themselves to Use Tools
• Meta AI: Toolformer publication

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