AI Paper Atlas

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

Towards Neural Programming Interfaces

Zachary Brown, Nathaniel Robinson, David Wingate, Nancy Fulda

NeurIPS 2020 · Large language models › Language, safety and interpretability · approach: Controlled text generation · 1 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
Train an NPI to induce the target word catGPT-2 · Cat inductiontarget in output: 0.00% -> 47.6%Improved
Train the cat-induction NPI on 691 rather than 70,227 text samples (ablation)cat-NPI · Cat inductiontarget in output: 47.6% -> 33.3%Mixed
Train an NPI to avoid catGPT-2 · Cat avoidancecat in output: 37.1% -> 11.0%Improved
Train an NPI to avoid a public figure's nameGPT-2 · Public figure avoidancename in output: 76.2% -> 52.4%Improved
Use NPI control rather than PPLM for cat inductionGPT-2 · Cat inductiontarget in output: 23.2% -> 48.8%Improved
Use NPI control rather than PPLM for cat avoidanceGPT-2 · Cat avoidancecat in output: 10.0% -> 11.2%Mixed
Train an NPI to avoid offensive termsGPT-2 · Offense avoidancetarget in output: 28.4% -> 17.6%Improved
Train an NPI to avoid racial slursGPT-2 · Racial slur avoidancetarget in output: 52.1% -> 0.5%Improved
Train an NPI to avoid gender slursGPT-2 · Gender slur avoidancetarget in output: 90.2% -> 10.3%Improved
Train an NPI to avoid a broad list of offensive termsGPT-2 · Offensive speech avoidancetarget in output: 88.4% -> 58.0%Improved
Train an NPI to induce shorter average wordsGPT-2 · Word length controlavg word length: 3.82 -> 2.90Improved
Train an NPI to induce longer average wordsGPT-2 · Word length controlavg word length: 3.82 -> 4.10Improved

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