AI Paper Atlas

Research map / Large language models / Language, safety and interpretability

Extensible Prompts for Language Models on Zero-shot Language Style Customization

Tao Ge, Jing Hu, Li Dong, Shaoguang Mao, Yan Xia, Xun Wang, Si-Qing Chen, Furu Wei

NeurIPS 2023 · Large language models › Language, safety and interpretability · approach: In-context learning analysis · 3 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
remove context-augmented learning (CAL) from X-Prompt (ablation)X-Prompt · 20 Users (OOD)perplexity: 28.5 -> 29.4Hurt
remove context-augmented learning (CAL) from X-Prompt (ablation)X-Prompt · 20 Users (OOD)accuracy: 38.6 -> 37.9Hurt
remove CAL from X-Prompt (ablation)X-Prompt · 20 Users (ID)perplexity: 30.8 -> 29.7Improved
remove CAL from X-Prompt (ablation)X-Prompt · 800 Users (ID)accuracy: 29.3 -> 29.9Improved
remove CAL from X-Prompt (ablation)X-Prompt · Human evaluation, open-ended generationoverall: 0.54 -> 0.35Hurt
remove CAL from X-Prompt (ablation)X-Prompt · Human evaluation, open-ended generationcontent: 0.69 -> 0.38Hurt
use X-Prompt instead of prompt tuningOPT-6.7b · 20 Users (OOD)perplexity: 29.5 -> 28.5Improved
use X-Prompt instead of prompt tuningOPT-6.7b · 20 Users (ID)accuracy: 37.8 -> 37.2Hurt
increase few-shot examples from 8 to 32No prompt baseline · 800 Users (ID)perplexity: 69.9 -> 62.7Improved
scale OPT from 350m to 6.7bX-Prompt · Human evaluation, overalloverall: 0.24 (+0.09) -> 0.54 (+0.24)Improved
scale OPT from 350m to 6.7bX-Prompt · Human evaluation, contentcontent: 0.30 (+0.08) -> 0.69 (+0.35)Improved
use X-Prompt instead of prompt tuning on styleOPT models · Human evaluation, stylestyle: 0.92 -> 0.83Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy