Li and Liang: prefix tuning for language generation

What changed

Kept model parameters frozen while learning task-specific continuous prefixes that later tokens can attend to during generation.

Scientific scope

Kept model parameters frozen while learning task-specific continuous prefixes that later tokens can attend to during generation. Experiments concern table-to-text generation and summarization. The method is separate from discrete prompting, input-only soft prompt tuning and low-rank weight updates.

Evidence

The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. Experiments concern table-to-text generation and summarization. The method is separate from discrete prompting, input-only soft prompt tuning and low-rank weight updates.

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.

• Prefix-Tuning: Optimizing Continuous Prompts for Generation
• ACL 2021: original prefix-tuning paper

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